This is the home page for NYU's Computer Science Theory seminar, organized by the NYU Theoretical Computer Science Group. We also host Discrete Mathematics talks organized by Jinyoung Park, so if you are looking for the Discrete Math seminar, you are in the right place!

Time and Location:

Thursday, 11AM
Warren Weaver Hall
251 Mercer Street
Room 317

To receive announcements about talks, you can sign up for the Theory Seminar mailing list here. You can also subscribe to our Google Calendar. For more information, or if you want to invite a speaker, please contact Christopher Musco.

Spring 2025

Jan. 9
  • Hamoon Mousavi (UC Berkeley)
  • Abstract: Finding an assignment to a set of variables (e.g., Boolean variables) that satisfies a given set of constraints is known as a constraint satisfaction problem (CSP). Maximizing the number of satisfied constraints can be framed as a problem of optimizing a multivariate polynomial over a discrete set (-1 and +1, in this case). If we relax the requirement that these variables commute, we enter the realm of noncommutative polynomial optimization (where the role of units -1 and +1 are played by unitary matrices). These resulting optimization problems, known as quantum constraint satisfaction problems, play a significant role in quantum information. They are also recognized by different names in various fields: physicists refer to them as nonlocal games, while computer scientists call them entangled multiprover interactive proofs.

    In the classical, commutative setting, we have a deep understanding of CSPs. For instance, we know which problems are NP-hard and how well we can approximate them. In this talk, I will explore the intriguing twists that arise in the quantum setting as we begin to understand quantum CSPs. Along the way we discuss a number of questions. Do we have a quantum PCP theorem? What about a quantum Unique Games Conjecture? How does the Majority Is Stablest theorem extend to the noncommutative world? Could the broader framework of noncommutative CSPs reveal new insights into classical CSPs? When considering quantum CSPs, could quantum approximation algorithms beat classical ones?

    Pardon the jargon in the latter questions. The talk will be fully self-contained and will not assume prior knowledge of quantum information or CSPs.
Jan. 16
  • Sayantan Sen (NUS)
  • Abstract: This work establishes a novel connection between PAC-learning of high-dimensional graphical models with efficient counting and sampling of graph structures using the well-studied online learning framework. The problem of efficiently counting and sampling graphical structures, such as spanning trees and acyclic orientations, has been a vibrant area of research in algorithms with significant advances over the past decade. We show that this rich algorithmic foundation can be leveraged to develop new algorithms for learning high-dimensional graphical models.

    We observe that if we apply the exponentially weighted average (EWA) or randomized weighted majority (RWM) algorithms (two algorithms extensively studied in the online learning literature) on a sequence of samples from an unknown distribution P with the log loss function, the average regret can bound the expected KL divergence between P and the predictions of EWA and RWM. This results in new sample complexity bounds for learning Bayes nets. Moreover, using this connection, we design the first efficient polynomial sample and time algorithm for sampling Bayes nets with a given chordal skeleton. Additionally, this approach gives a new algorithm of learning tree-structured distributions, different from the well-known Chow-Liu algorithm.

    This is a joint work with Arnab Bhattacharyya, Sutanu Gayen, Philips George John and N. V. Vinodchandran, and preprint is available at https://arxiv.org/abs/2405.07914.
Jan. 23
  • Jalaj Upadhyay (Rutgers)
  • Abstract: Differential privacy is now considered the de facto notion of privacy; however, there is still a wide gap between the theory and practice of differential privacy. This talk will discuss the fundamental reasons for this discrepancy using continual counting as an example. We will briefly overview the problem and discuss recent advances in differentially private continual observation that achieves fine-grained error bounds. Time permitting, we will also discuss some of the implications of these results in both real-world applications and abstract mathematics.

    This talk is based on joint work with Monika Henzinger and Sarvagya Upadhyay
Jan. 30
  • Uma Girish (Columbia)
  • Abstract: We study the power of entanglement in communication complexity. We consider classical communication models that are equipped with a limited number of qubits of entanglement, i.e., models where Alice and Bob share entanglement and can exchange classical messages that depend on measurement outcomes of their shared state. We demonstrate in a number of settings that reducing the entanglement even by a small factor can exponentially increase the communication complexity. Among several results, we show that for every \(k >= 1\):

    1. Quantum simultaneous protocols with \(O(k^5 \log^3 N)\) qubits of entanglement can exponentially outperform classical two-way protocols with \(O(k)\) qubits of entanglement. This improves the state-of-the-art separations between quantum and classical communication complexity.

    2. Classical simultaneous protocols with \(O(k \log N)\) qubits of entanglement can exponentially outperform quantum simultaneous protocols with \(O(k)\) qubits of entanglement.

    This is based on a joint work with Srinivasan Arunachalam.
Feb. 6
  • Shi Li (Nanjing University)
  • Abstract: In this talk, I will give our recent results on the problem of assigning items to agents so as to maximize the weighted Nash Social Welfare (NSW) under additive and submodular valuations. Obtaining a constant approximation algorithm for problem is an open problem in the field that has recently attracted considerable attention.

    For the additive valuation case, we give an \(e^{1/e} + \epsilon \approx 1.445 + \epsilon\) approximation algorithm for the problem, which matches the best known result for the unweighted case. Both our algorithm and analysis are simple. We solve a natural configuration LP for the problem, and obtain the allocation of items to agents using a randomized version of the Shmoys-Tardos rounding algorithm developed for unrelated machine scheduling problems.

    For the submodular case, we give the first \(O(1)\)-approximation algorithm, also solving the open problem in the affirmative. Our algorithm is also based on the natural configuration LP for the problem. Our rounding algorithm is similar to that of [Li, 2025] developed for the unrelated machine scheduling problem to minimize weighted completion time.

    This talk is based on two papers: [Feng-Li, 2024] and [Feng-Hu-Li-Zhang, 2025].
Feb. 13
  • Tal Herman (Weizmann Institute of Science)
  • Abstract: Learning about an unknown distribution using i.i.d. samples is a basic challenge in many scientific disciplines. There are many tasks we could consider: from estimating statistical quantities, such as the distribution’s support size or Shannon entropy, to supervised or unsupervised learning. In many cases, these tasks require drawing a large number of samples from the distribution and significant computational resources.

    How about verifying the results of such an analysis? Can an untrusted analyst provide a proof of correctness for the results? We are interested in proofs that can be verified using fewer resources (samples and running time) than what is needed for performing the task on one's own.

    The talk will describe a new line of work on constructing efficient interactive proof systems for verifying properties of unknown distributions, demonstrating the power of such proofs for a very rich family of distribution properties.

    The talk is based on joint works with prof. Guy Rothblum.
Feb. 20
No meeting this week.

Feb. 27
  • Junpei Komiyama (NYU Stern)
  • Abstract: We consider the fixed-budget best arm identification problem, where the goal is to find the arm of the largest mean with a fixed number of samples. This problem can model adaptive A/B testing and can be used as a search component for efficient Monte Carlo tree search. It is known that the probability of misidentifying the best arm is exponentially small to the number of rounds. However, limited characterizations have been discussed on this value's exponent. In this paper, we characterize the minimax optimal rate and discuss algorithms that achieve this rate. After these technical considerations, we briefly discuss relation to purely explorative problems, such as best action identification in Monte Carlo tree search, which is increasingly popular in test time inference in large language models.
Mar. 6
  • Shivam Nadimpalli (Massachusetts Institute of Technology)
  • Abstract: Given an intersection of (possibly infinitely many) halfspaces at bounded distance from the origin, we show that it can be sparsified, i.e. approximated (under the Gaussian distribution) by an intersection of a halfspaces where the number of halfspaces depends only on the desired accuracy. This yields efficient algorithms for learning, tolerant testing, and volume estimation of convex sets of bounded width.

    Our result follows from a more general sparsification lemma for Gaussian processes, which relies on Talagrand's majorizing measures theorem. As another consequence, we obtain a "junta theorem" for norms over Gaussian space: Every norm over R^n can be multiplicatively approximated (under the Gaussian measure) by a norm that depends on only a constant number of coordinates.

    Based on joint work with Anindya De, Ryan O'Donnell, and Rocco Servedio: https://arxiv.org/abs/2411.14664.
Mar. 13
No talk.

Mar. 20
  • Thuy-Duong (June) Vuong (Berkeley)
  • Abstract: Learning to sample is a central task in generative AI: the goal is to generate (infinitely many more) samples from a target distribution \(\mu\) given a small number of samples from \(\mu.\) It is well-known that traditional algorithms such as Glauber or Langevin dynamics are highly inefficient when the target distribution is multimodal, as they take exponential time to converge from a worst case start, while recently proposed algorithms like denoising diffusion (DDPM) require information that is computationally hard to learn. In this talk, we propose a novel and conceptually simple algorithmic framework to learn multimodal target distributions by initializing traditional sampling algorithms at the empirical distribution. As applications, we show new results for two representative distribution families: Gaussian mixtures and Ising models. When the target distribution \(\mu\) is a mixture of \(k\) well-conditioned Gaussians, we show that the (continuous) Langevin dynamics initialized from the empirical distribution over \(\tilde{O}(k/\epsilon^2)\) samples, with high probability over the samples, converge to \(mu\) in \(\tilde{O}(1)\)-time; both the number of samples and convergence time are optimal. When \(\mu\) is a low-complexity Ising model, we show a similar result for the Glauber dynamics with approximate marginals learned via pseudolikelihood estimation, demonstrating for the first time that such low-complexity Ising models can be efficiently learned from samples.

    Based on joint work with Frederic Koehler and Holden Lee.
Mar. 27
  • Nairen Cao (New York University)
  • Abstract: Correlation Clustering is one of the most extensively studied graph clustering problems. The input is a complete graph where each edge is labeled either “+” or “−,” and the objective is to partition the vertex set into an arbitrary number of clusters to minimize the sum of (i) the number of “+” edges spanning different clusters and (ii) the number of "−" edges within the same cluster.

    To establish a simple yet powerful framework for Correlation Clustering—akin to those used in typical approximation algorithms—we introduce the Clustering LP, a strong linear program that we conjecture closely captures the approximability of the problem. We demonstrate the strength of the Clustering LP by presenting a simple rounding algorithm that achieves a 1.437-approximation, surpassing previous results.

    This work is a joint collaboration with Vincent Cohen-Addad, Shi Li, Euiwoong Lee, David Rasmussen Lolck, Alantha Newman, Mikkel Thorup, Lukas Vogl, Shuyi Yan, and Hanwen Zhang.

Apr. 3
  • Mursalin Habib (Rutgers)
  • Abstract: A function that maps n-bit strings to N-bit strings is called an isometric embedding of the n-dimensional Hamming metric space to the N-dimensional edit metric space if the Hamming distance between any pair of input strings equals the edit distance between the corresponding output strings. The rate of such an embedding is defined as the ratio n/N.

    It is well known in the literature how to construct isometric embeddings with a rate of 1/logn. However, achieving even near-isometric embeddings with a positive constant rate has remained elusive until now.

    In this talk, I will go over our recent work giving the first-ever constant rate isometric embedding of the Hamming metric into the edit metric. In particular, I will describe an isometric embedding with a rate of 1/8 by discovering connections to synchronization strings, which were studied in the context of insertion-deletion codes (Haeupler-Shahrasbi [JACM'21]).

    As an immediate consequence of our constant rate isometric embedding, we improve known conditional lower bounds for the closest pair problem and the discrete 1-center problem in edit metric and NP-hardness of approximation results for clustering problems and the Steiner tree problem in the edit metric, but now with optimal dependency on the dimension.

    At a technical level, we introduce a framework for obtaining high-rate isometric embeddings using novel objects called misaligners. We speculate that, with sufficient computational resources, our framework could potentially yield isometric embeddings with a rate of 1/5. We also prove that any isometric embedding must involve interleaving the input bits with bit strings that do not depend on the input. This structural result allows us to prove upper bounds on the rate of any isometric embedding.

    This is joint work with Sudatta Bhattacharya, Elazar Goldenberg, Sanjana Dey, Bernhard Haeupler, Karthik C.S., and Michal Koucký.
Apr. 10
  • Ben Moseley (Carnegie Mellon University)
  • Abstract: This talk explores the emerging area of algorithms with predictions, also known as learning-augmented algorithms. These methods incorporate machine-learned predictions into algorithmic design, allowing the algorithms to adapt to input distributions and achieve improved performance in runtime, space, or solution quality. The presentation will highlight recent advances in leveraging predictions to enhance the efficiency of optimization algorithms and dynamic graph algorithms.

    A key focus will be on achieving "instance-optimal" performance—where algorithms excel when predictions are accurate—while ensuring graceful degradation when predictions are imperfect. Through examples such as bipartite matching, the talk will demonstrate the transformative potential of this approach to significantly improve algorithmic efficiency.
Apr. 17
  • Akbar Rafiey (New York University)
  • Abstract: In this talk, we explore continuous extensions of combinatorial optimization problems and their applications in unsupervised learning. Combinatorial problems over discrete domains are inherently incompatible with gradient-based optimization and deep learning architectures. We address these challenges for two broad classes of problems: those over permutations (e.g., TSP, CutWidth) and set functions.

    We introduce continuous, differentiable extensions that are polynomial-time computable and come with rounding guarantees. These properties enable the design of optimization algorithms akin to the Frank-Wolfe method for permutation-based problems. Moreover, our extensions integrate naturally with the backpropagation framework, allowing them to serve as loss functions in data-driven, unsupervised learning pipelines.
Apr. 24
  • Jesse Goodman (UT Austin)
  • Abstract: Given a sequence of \(N\) independent sources of randomness \(X_1, \ldots, X_N\), how many of them must be *good* (i.e., contain some entropy) in order to extract a uniformly random string? This question was first raised by by Chattopadhyay, Goodman, Goyal and Li (STOC'20), motivated by applications in cryptography, and by the unreliable nature of real-world randomness. There, it was shown how to extract uniform bits with very low error, even when just \(K = N^{0.5}\) of the sources are good. In a follow-up work at FOCS'21, the number of good sources required was significantly improved to \(K = N^{0.01}\).

    In this talk, I will discuss our new work that achieves \(K = 3\).

    Our key ingredient is a near-optimal explicit construction of a powerful new pseudorandom primitive, called a "leakage-resilient extractor (LRE) against number-on-forehead (NOF) protocols." Our LRE can be viewed as a significantly more robust version of classical independent source extractors, and resolves an open question by Kumar, Meka, and Sahai (FOCS'19) and Chattopadhyay, Goodman, Goyal, Kumar, Li, Meka, and Zuckerman (FOCS'20). Our LRE construction is based on a simple new connection we discover between multiparty communication complexity and "non-malleable" extractors, which shows that such extractors exhibit strong average-case lower bounds against NOF protocols.

    Based on joint work with Eshan Chattopadhyay (STOC'25, to appear).
May 1
  • Sumegha Garg (Rutgers)
  • Abstract: In this talk, we will introduce a new notion of information complexity for one-pass and multi-pass streaming problems, and use it to prove memory lower bounds for the coin problem. In the coin problem, one sees a stream of \(n\) i.i.d. uniform bits and one would like to compute the majority (or sum) with constant advantage. We show that any constant pass algorithm must use \(\Omega(log n)\) bits of memory. This information complexity notion is also useful to prove tight space complexity for the needle problem, which in turn implies tight bounds for the problem of approximating higher frequency moments in a data stream.

    Joint works with Mark Braverman, Qian Li, Shuo Wang, David P. Woodruff and Jiapeng Zhang.
May 8
  • Prasanna Ramakrishnan (Stanford)
  • Abstract: In 1785, Condorcet established a frustrating property of elections and majority rule: it is possible that, no matter which candidate you pick as the winner, a majority of voters will prefer someone else. You might have the brilliant idea of picking a small set of winners instead of just one, but how do you avoid the nightmare scenario where a majority of the voters prefer some other candidate over all the ones you picked? How many candidates suffice to appease a majority of the voters? In this talk, we will explore this question. Along the way, we will roll some dice — both because the analysis involves randomness and because of a connection to the curious phenomenon of intransitive dice, that has delighted recreational and professional mathematicians alike ever since Martin Gardner popularized it in 1970.

    Based on joint work with Moses Charikar, Alexandra Lassota, Adrian Vetta, and Kangning Wang
May 12
  • Jie Xue (NYU Shanghai)
  • Abstract: Many fundamental graph problems can be formulated as hitting certain small patterns in a given graph using a minimum number of vertices. This motivates the F-Hitting problem. Specifically, for a finite set F of graphs, the F-Hitting problem takes a graph G and a parameter k as input and aims to compute a set S of at most k vertices such that G-S does not contain any graph in F as a subgraph. In this talk, we shall focus on the F-Hitting problem and discuss a recent algorithmic framework for the problem. The framework yields subexponential parameterized algorithms for F-Hitting on a broad family of graph classes.
June 12
  • Joel Tropp (Caltech)
  • Abstract: On the real line, a random walk that can only move in the positive direction is very unlikely to remain close to its starting point. After a fixed number of steps, the left tail has a Gaussian profile, under minimal assumptions. Remarkably, the same phenomenon occurs when we consider a positive random walk on the cone of positive-semidefinite matrices. After a fixed number of steps, the minimum eigenvalue is also described by a Gaussian model.

    This talk introduces a new way to make this intuition rigorous. The methodology provides the solution to an open problem in computational mathematics about sparse random embeddings. The presentation is targeted at a general mathematical audience.
Sept. 19
  • Jeremy Fineman (Georgetown)
  • Abstract: This talk presents a new randomized algorithm for the problem of single-source shortest path on directed graphs with real (both positive and negative) edge weights. Given an input graph with n vertices and m edges, the algorithm completes in \(\tilde{O}(mn^{8/9})\) time with high probability. Whereas there has been significant progress for the case of integer-weighted graphs, this result constitutes the first asymptotic improvement over the classic \(O(mn)\)-time algorithm for the case of real weights.
Sept. 20 (Dept. Seminar at Tandon CSE)

  • John Iacono (Universite libre de Bruxelles)
  • Time/Location: 11am-12pm, 370 Jay Street, 8th Floor, Room 825
  • Abstract: We consider the problem of sorting n items, given the outcomes of m pre-existing comparisons. We present a simple, natural deterministic algorithm that runs in \(O(m+ \log T)\) time and does \(O(\log T)\) comparisons, where \(T\) is the number of total orders consistent with the pre-existing comparisons.

    Our running time and comparison bounds are best possible up to constant factors, thus resolving a problem that has been studied intensely since 1976 (Fredman, Theoretical Computer Science). The best previous algorithm with a bound of \(O(\log T)\) on the number of comparisons has a time bound of \(O(n^{2.5})\) and is more complicated. (A recent independent and concurrent work by Van der Hoog and Rutschmann implies an \(O(n^{2.371552})\)-time algorithm with \(O(\log T)\) comparisons.)

    Our algorithm combines three classic algorithms: topological sort, heapsort with the right kind of heap, and efficient insertion into a sorted list. It outputs the items in sorted order one by one. As a result, it can be modified to solve the important and more general top-k sorting problem: Given k and the outcomes of some pre-existing comparisons, output the smallest k items in sorted order. The modified algorithm solves the top-k sorting problem in minimum time and comparisons, to within constant factors.

    Joint work with Bernhard Haeupler, Richard Hladík, Vaclav Rozhon, Robert Tarjan, and Jakub Tětek
Sept. 26
  • Jackie Baek (NYU Stern)
  • Abstract: Exploration is often necessary in online learning to maximize long-term rewards, but it comes at the cost of short-term “regret.” We study how this cost of exploration is shared across multiple groups. For example, in a clinical trial setting, patients who are assigned a suboptimal treatment effectively incur the cost of exploration. When patients are associated with natural groups on the basis of, say, race or age, it is natural to ask whether the cost of exploration borne by any single group is “fair.” So motivated, we introduce the “grouped” bandit model. We leverage the theory of axiomatic bargaining, and the Nash bargaining solution in particular, to formalize what might constitute a fair division of the cost of exploration across groups. On one hand, we show that any regret-optimal policy strikingly results in the least fair outcome: such policies will perversely leverage the most “disadvantaged” groups when they can. More constructively, we derive policies that are optimally fair and simultaneously enjoy a small “price of fairness.” We illustrate the relative merits of our algorithmic framework with a case study on contextual bandits for warfarin dosing where we are concerned with the cost of exploration across multiple races and age groups.

    Joint work with Vivek Farias.
Oct. 3
  • Andreas Maggiori (Columbia)
  • Abstract: We introduce a new problem of portfolio optimization using stochastic information. In a setting where there is some uncertainty, we ask how to best select \(k\) potential solutions, with the goal of optimizing the value of the best solution. More formally, given a combinatorial problem \(\Pi\), a set of value functions \(\mathcal{V}\) over the solutions of \(\Pi\), and a distribution \(\mathcal{D}\) over \(\mathcal{V}\), our goal is to select \(k\) solutions of \(\Pi\) that maximize or minimize the expected value of the best of those solutions.

    For a simple example, consider the classic knapsack problem: given a universe of elements each with unit weight and a positive value, select a set of \(r\) elements maximizing the total value. Now suppose that each element's weight comes from a (known) distribution. How to best select \(k\) different solutions, so that one of them is likely to be near optimal? In this work we tackle this basic problem, and generalize it to the setting where the underlying set system forms a matroid. On the technical side, it is clear that the candidate solutions we select must be diverse and anti-correlated, however, it is not clear how to do so efficiently. Our main result is a polynomial-time algorithm that computes "anti-concentrated" solutions and constructs a portfolio within a constant factor of the optimal.

    This is joint work with: Marina Drygala, Silvio Lattanzi, Miltiadis Stouras, Ola Svensson and Sergei Vassilvitskii
Oct. 10
  • Mitali Bafna (Massachusetts Institute of Technology)
  • Abstract: The PCP Theorem [FGLSS,AS,ALMSS'92] is a cornerstone of theoretical computer science, with many applications in hardness of approximation, cryptography and interactive protocols. Thus, building size-efficient probabilistically checkable proofs (PCPs) with low soundness has been an important question in complexity theory. We construct 2-query, quasi-linear size PCPs with arbitrarily small constant soundness, improving upon Dinur's 2-query quasi-linear size PCPs with soundness \(1-\Omega(1)\). As an immediate corollary, we get that under the exponential time hypothesis, for all \(\epsilon>0\) no approximation algorithm for 3-SAT can obtain an approximation ratio of \(7/8+\epsilon\) in time \(2^{n/\log^C n}\), where \(C\) is a constant depending on \eps. Our result builds on a recent line of independent works, by Bafna, Lifshitz, Minzer, and Dikstein, Dinur, Lubotzky that show the existence of linear size direct product testers with small soundness.

    The main new ingredient in our proof is a technique that embeds a given 2-CSP (2-variate constraint-satisfaction problem) into a 2-CSP on a prescribed graph, provided that the latter is a graph underlying a sufficiently good high-dimensional expander (HDX). Towards this end, we show a novel connection between fault-tolerant distributed computing and PCPs, more precisely from the literature of the almost everywhere agreement problem defined in the work of Dwork, Peleg, Pippenger, and Upfal (1986). We instantiate this connection by showing that graphs underlying HDXs admit routing protocols that are tolerant to adversarial edge corruptions. The latter result also improves upon the state of the art in constructions of fault-tolerant sparse networks.

    Based on joint work with Dor Minzer and Nikhil Vyas and Appendix by Zhiwei Yun.
Oct. 17
  • Sanjeev Khanna (University of Pennsylvania)
  • Abstract: The fully dynamic matching problem involves efficiently maintaining a near-optimal matching in a graph undergoing edge insertions and deletions. Despite significant effort, the goal of designing highly efficient fully dynamic matching algorithms has remained elusive. A promising natural avenue for obtaining faster algorithms is to maintain a matching sparsifier, that is, a sparse subgraph that approximately preserves large matchings in every induced subgraph of the original graph. The success of this approach hinges on a positive resolution of two challenges: showing the existence of sparse matching sparsifiers and designing efficient algorithms for constructing them.

    As it turns out, the first challenge is intimately connected to the question of determining the density of Ruzsa-Szemerédi (RS) graphs, namely, graphs whose edges can be partitioned into induced matchings of linear size. However, even an optimistic resolution of the RS graph density question would still leave open the second challenge. An exciting recent insight has highlighted a relaxed notion of RS graphs, called ordered RS graphs, that surprisingly allows folding both challenges above into a single combinatorial question on the density of such graphs.

    This talk will explore the interplay between these ideas, and will show how this interplay has led to a conditional dream result for fully dynamic matching.

Oct. 24
  • Tom Kelly (Georgia Tech)
  • Abstract: A graph H embeds in a graph G if G contains a subgraph isomorphic to H, and it decomposes G if the edges of G can be partitioned into subgraphs isomorphic to H. Questions about when a graph embeds in or decomposes another are central in combinatorics. “Dirac-type” embedding results address minimum-degree conditions to ensure an embedding of some graph. Block designs, a fundamental object of Design Theory, are decompositions of complete graphs.

    In this talk, we will discuss robustness of embeddings and decompositions. For example, given a hypergraph of large minimum degree, we will discuss the threshold for a random subhypergraph to have a perfect matching or Hamilton cycle. We will also discuss the threshold for constructing block designs using only a random selection of blocks. All of these results utilize the recent Park–Pham Theorem or one of its variants. A crucial notion for this is that of the spreadness of a certain type of probability distribution.
Oct. 31
  • Zongrui Zou (Nanjing University)
  • Abstract: In this talk, we show how to address the challenge of differential privacy in the context of graph cuts, specifically focusing on the minimum \(k\)-cut and multiway cut problems. We introduce edge-differentially private algorithms that achieve nearly optimal performance for these problems. For the multiway cut problem, we first provide a private algorithm with a multiplicative approximation ratio that matches the state-of-the-art non-private algorithm. We then present a tight information-theoretic lower bound on the additive error, demonstrating that our algorithm on weighted graphs is near-optimal for constant \(k\). For the minimum \(k\)-cut problem, our algorithms leverage a known bound on the number of approximate \(k\)-cuts, resulting in a private algorithm with optimal additive error \(O(k\log n)\) for fixed privacy parameter. We also establish a information-theoretic lower bound that matches this additive error. Additionally, we give an efficient private algorithm for \(k\)-cut even for non-constant \(k\), including a polynomial-time 2-approximation with an additive error of \(\widetilde{O}(k^{1.5})\).
Nov. 7
  • Valerie King (University of Victoria)
  • Abstract: Byzantine agreement is a fundamental problem in distributed computing which was formulated over 50 years ago to model the coordination of machines in a network which experience arbitrary faults. More recent applications like blockchain require consideration of a wider range of models and problems. In this talk we'll explore a novel model introduced by Augustine et al. in DISC 2024 in which the goal is for a set of possibly corrupt machines to cooperate in order to efficiently share the cost of learning external data. Unlike the standard Byzantine agreement model, in this context there is a surprisingly efficient algorithm which works in a permissionless system provided there is a known fixed constant fraction of non-corrupted agents.
Nov. 14
  • Tyler Chen (New York University)
  • Abstract: An important task in scientific machine learning is operator learning, where we aim to learn an approximation of an operator (e.g. the solution operator to a given differential equation) from input/output data (e.g. forcing function/solution pairs). This approximation can then be used to make predictions about future input data cheaply. Hierarchical matrices form a natural hypothesis class for many operator learning problems arising from physical systems, but a theoretical understanding of how to construct provably accurate estimators from data has been elusive. We describe a simple data-generation process and efficiently implementable algorithm for producing a near-optimal hierarchically off-diagonal low-rank (HODLR) approximation to an arbitrary matrix. In particular, we show that we can obtain an HODLR approximation competitive with the best possible HODLR approximation using a nearly-optimal number of data-points. Our analysis relies on a new perturbation bound for randomized low-rank approximation that may be of broader interest.
Nov. 21
  • Renfei Zhou (Carnegie Mellon University)
  • Abstract: The seminal work "Succincter" by Pătraşcu presents a way to store an "augmented B-tree" with only two bits of redundancy while supporting queries efficiently. It is a generic and powerful tool for designing static succinct data structures. We extend "Succincter" to support dynamic operations, achieving the same space bounds as the static "Succincter" and the optimal time bounds, enabling numerous applications. Our technique addresses a key challenge in dynamic succinct data structures: managing variable-size components within a contiguous piece of memory.
Nov. 28
  • No meeting (Thanksgiving)
Dec. 5
  • Jingxun Liang (Carnegie Mellon University)
  • Abstract: The dictionary problem involves maintaining a set \(S \subset [U]\) of n keys under insertions, deletions, and membership queries focusing on efficiency in both time and space. This fundamental problem has been studied extensively for six decades since 1953. Recently, a series of works has established the optimal time-space trade-off for dictionaries. In this talk, I will describe the tight lower bound result that any dictionary with an operational time \(t\) must incur a redundant \(\Omega(\log^{(t)} n)\) bits per key in space.
Dec. 10 (Special Seminar)

  • Nati Linial (The Hebrew University of Jerusalem)
  • Time/Location: 11am-12pm, 517 Warren Weaver Hall
  • Abstract: Paraphrasing the title of Riemann’s famous lecture of 1854 I ask: What is the most rudimentary notion of a geometry? A possible answer is a path system: Consider a finite set of "points" \(x_1,…,x_n\) and provide a recipe how to walk between \(x_i\) and \(x_j\) for all \(i\neq j\), namely decide on a path \(P_{ij}\), i.e., a sequence of points that starts at \(x_i\) and ends at \(x_j\), where \(P_{ji}\) is \(P_{ij}\), in reverse order. The main property that we consider is consistency. A path system is called consistent if it is closed under taking subpaths. What do such systems look like? How to generate all of them? We still do not know. One way to generate a consistent path system is to associate a positive number \(w_{ij}>0\) with every pair and let \(P_{ij}\) be the corresponding \(w\)-shortest path between \(x_i\) and \(x_j\). Such a path system is called metrical. It turns out that the class of consistent path systems is way richer than the metrical ones.

    My main emphasis in this lecture is on what we don't know and wish to know, yet there is already a considerable body of work that we have done on the subject.

    The new results that I will present are joint with my student Daniel Cizma as well as with Maria Chudnovsky.
Dec. 12
  • Xiaoyu He (Georgia Tech)
  • Abstract: It is a fundamental question in Ramsey theory to determine the smallest possible independence number of an H-free hypergraph on n vertices. In the case of graphs, the problem was famously solved for \(H=K_3\) by Kim and for \(H=K_4\) (up to a logarithmic factor) by Mattheus-Verstraete in 2023. Even \(C_4\) and \(K_5\) remain wide open.

    We study the problem for 3-uniform hypergraphs and conjecture a full classification: the minimum independence number is poly(n) if and only if H is contained in the iterated blowup of the single-edge hypergraph. We prove this conjecture for all H with at most two tightly connected components.

    Based on joint work with Conlon, Fox, Gunby, Mubayi, Suk, Verstraete, and Yu.
Jan. 11
  • Radu-Cristian Curticapean (IT University of Copenhagen)
  • Abstract: In recent years, homomorphism counts between graphs were studied in different subfields within theoretical computer science: From a complexity-theoretic point of view, it was shown that various counting problems can be understood well by expressing them as linear combinations of homomorphism counts. Such linear combinations enjoy the exceptional and very useful "monotonicity property" of being exactly as hard to compute as their hardest terms. Since the complexity of computing the individual terms, i.e., of counting homomorphisms, is reasonably well understood, this strategy allowed the community to make significant progress in the study of the complexity of various counting problems arising in database theory and network sciences.

    From a more abstract perspective, the numbers of homomorphisms between graphs encode valuable information. For example, a classical result by Lovász shows that homomorphism counts from all graphs into a given graph G identify G up to isomorphism. Restricting the class of graphs from which homomorphisms are counted gives rise to natural relaxations of graph isomorphism that are connected to the expressiveness of logics.

    In this self-contained talk, we survey recent developments in the study of computational and abstract properties of homomorphism counts.
Jan. 25
  • Kunal Marwaha (University of Chicago)
  • Abstract: The holy grail of quantum computing is a practical use for today's machines. A popular suggestion is that quantum computers can approximately solve optimization problems better than classical computers, despite little theoretical evidence. I take this claim seriously, analyzing and comparing average-case algorithms on CSPs of large (but fixed) clause density. It turns out that both algorithms and obstructions from spin glass theory naturally transfer to sparse CSPs, culminating in an optimal algorithm among all bounded-fanout quantum and classical circuits of depth up to \(\epsilon · \log n\). This talk surveys several recent works, especially BFMVZ21, JMSS22, and CHM23.
Feb. 1
  • Yang P. Liu (Institute for Advanced Study)
  • Abstract: We consider the sparsification of sums \(F : \mathbb{R}^n \to \mathbb{R}_+\) where \(F(x) = f_1(\langle a_1,x \rangle) + ... + f_m(\langle a_m,x \rangle)\) for vectors \(a_1, ..., a_m \in \mathbb{R}^n\) and functions \(f_1, ... ,f_m : \mathbb{R} \to \mathbb{R}_+\). We show that \((1+\epsilon)\)-approximate sparsifiers of \(F\) with support size \(\frac{n}{\epsilon^2} (\log \frac{n}{\epsilon})^{O(1)}\) exist whenever the functions \(f_1, ... ,f_m\) are symmetric, monotone, and satisfy natural growth bounds. Additionally, we give efficient algorithms to compute such a sparsifier assuming each \(f_i\) can be evaluated efficiently. Our results generalize the classic case of \(\ell_p\) sparsification, where \(f_i(z) = |z|^p\), for \(p \in (0, 2]\), and give the first near-linear size sparsifiers in the well-studied setting of the Huber loss function and its generalizations, e.g., \(f_i(z) = \min\{|z|^p, |z|^2\}\) for \(0 ＜ p \leq 2\). Our sparsification algorithm can be applied to give near-optimal reductions for optimizing a variety of generalized linear models including \(\ell_p\) regression for \(p \in (1, 2]\) to high accuracy, via solving \((\log n)^{O(1)}\) sparse regression instances with \(m \le n(\log n)^{O(1)}\), plus runtime proportional to the number of nonzero entries in the vectors \(a_1, \dots, a_m\). Based on joint work with Arun Jambulapati, James Lee, and Aaron Sidford.
Feb. 8
  • Ilias Zadik (Yale University)
  • Abstract: In this talk, we will present a recent result that if a Boolean function exhibits a sharp threshold at an arbitrary density then it immediately implies a bounded depth circuit lower bounds for computing this function on average. By leveraging the rich literature of sharp thresholds, this method implies multiple new average-case circuit lower bounds. Time permitted, we will discuss new AC0 lower bounds for the random 2-SAT problem at the satisfiability threshold, the k-clique decision problem in random graphs, and also the planted clique model from the literature of statistical estimation. Joint work with David Gamarnik and Elchanan Mossel.
Feb. 15
  • Kevin Schewior (University of Southern Denmark)
  • Abstract: Consider the following online problem: There is a gambler who receives a finite stream of independent draws from a probability distribution on the positive real numbers. The gambler can stop the stream at any point and receive the current value as final payoff. When the distribution is known, it is known that the gambler can achieve an expected payoff of at least roughly 0.745 times the expected maximum of the stream, the value of the “prophet”. In this talk, we review this result and discuss variants in which the distribution is not (or partially) known, in which the gambler only receives limited information about the values, and in which the goal is not just to stop a single time.
Feb. 22
  • Ainesh Bakshi (Massachusetts Institute of Technology)
  • Abstract: A quantum system of \(n\) interacting particles at thermal equilibrium can be described using a polynomial number of parameters, which correspond to the strengths of the forces between particles. A natural open question has been whether these parameters can be estimated efficiently by running experiments on the system. We resolve this question by giving the first polynomial-time algorithm for this task. This improves over prior work, which uses polynomially many samples from the system, but requires exponential time. In this talk, I will introduce the problem and describe some of the key ideas behind our algorithm. No background in quantum information is required.

    Based on joint work with Allen Liu, Ankur Moitra and Ewin Tang.
Feb. 29
  • Abhinav Bhardwaj (Yale University)
  • Abstract: Perturbation theory is developed to analyze the impact of noise on data and has been an essential part of numerical analysis. Recently, it has played an important role in designing and analyzing matrix algorithms. One of the most useful tools in this subject, the Davis-Kahan sine theorem, provides an \(\ell_2\) error bound on the perturbation of the leading singular vectors (and spaces). We focus on the case when the signal matrix has low rank and the perturbation is random, which occurs often in practice. In an earlier paper, O'Rourke, Wang, and the second author showed that in this case, one can obtain an improved theorem. In particular, the noise-to-gap ratio condition in the original setting can be weakened considerably. In the current paper, we develop an infinity norm version of the O'Rourke-Vu-Wang result. The key ideas in the proof are a new bootstrapping argument and the so-called iterative leave-one-out method, which may be of independent interest. Applying the new bounds, we develop new, simple, and quick algorithms for several well-known problems, such as finding hidden partitions and matrix completion. The core of these new algorithms is the fact that one is now able to quickly approximate certain key objects in the infinity norm, which has critical advantages over approximations in the \(\ell_2\) norm, Frobenius norm, or spectral norm.

    Joint work with Prof. Van Vu.
Mar. 7
  • Matthew Jenssen (King's College London)
  • Abstract: The classical sphere packing problem asks: what is the densest possible arrangement of identical, non-overlapping spheres in \(\mathbb{R}^d\)?

    I will discuss a recent proof that there exists a sphere packing with density at least \( (1-o(1))\frac{d \log d}{2^{d+1}}. \) This improves upon previous bounds by a factor of order \(\log d\) and is the first improvement by more than a constant factor to Rogers' bound from 1947.

    This is joint work with Marcelo Campos, Marcus Michelen and Julian Sahasrabudhe.
Mar. 14
  • Radoslav Fulek (New York University)
  • Abstract: We give a gentle introduction to research problems related to deciding whether the self-intersections of a given polygon can be eliminated by its infinitesimal perturbation. In computer science, the problem is known as the recognition of weakly simple polygons. Weakly simple polygons naturally generalize to the so-called ``weak embeddings of graphs'''. This scenario arises in applications in clustering, cartography, and graph visualization, where nearby vertices and edges are often bundled to the same point or overlapping arcs due to data compression, graph semantics, or low resolution. The recent advances in the area led to the resolution of several seemingly unrelated long standing open problems in computational geometry, such as the complexity status of the embeddability of simply connected polyhedra in \(\mathbb{R}^3\) and clustered planarity. We cover recent developments and open problems in the area, mainly, the problem of deciding whether a pair of polygonal chains can be made disjoint by infinitesimal perturbations.

    Based on joint work with Hugo Akitaya and Csaba Toth.
Mar. 21
  • No Meeting -- Spring Break.
Mar. 28
  • Yuri Faenza (Columbia University)
  • Abstract: Gale and Shapley's family of stable matchings enjoys a rich mathematical structure, that has been exploited over the years to construct efficient algorithms. In this talk, we investigate alternatives to stable matchings that rely on the concept of internal stability, a notion introduced for abstract games by von Neumann and Morgenstern and motivated by the need of finding a set of mutually compatible solutions. The set of stable matchings is internally stable (IS). However, the class of IS sets is much richer, for an IS set of matchings may also include unstable matchings and/or exclude stable ones. In this paper, we focus on two families of IS sets of matchings: von Neumann-Morgenstern (vNM) stable and internally closed. We study structural and algorithmic questions around those concepts in both the marriage (ie, bipartite) and roommate (ie, non-bipartite) model. Our results imply that, in the marriage model, internally closed and vNM stable sets are an alternative to stable matchings that are as tractable as stable matchings themselves, a fairly rare occurrence in the area. En route to our results, we present classical algebraic structures associated with stable matchings, and some of their generalizations that we derive in our work.

    Based on joint work with Cliff Stein, Jia Wan, and Xuan Zhang.

Apr. 4
  • Guy Moshkovitz (City University of New York)
  • Abstract: While for matrices there is one notion of rank with many equivalent definitions, for tensors -- or higher-dimensional matrices -- there are many inequivalent notions. The interplay between these notions plays a central role in the context of the structure-vs-randomness paradigm, and is closely related to important questions in additive combinatorics (e.g. Polynomial Freiman–Ruzsa) and computer science (e.g. Goldreich-Levin theorems), to name a few. In this talk I will discuss recent (exciting!) progress in the study of tensor ranks from the last few years.
Apr. 11
  • Dingding Dong (Harvard University)
  • Abstract: We discuss recent work on the enumeration of k-SAT functions (boolean functions that can be represented as k-SAT formulae). For any fixed constant k, we show that almost all k-SAT functions on n Boolean variables are unate, i.e., monotone after first negating some variables. This resolves a conjecture by Bollobás, Brightwell, and Leader from 2003. Joint work with József Balogh, Bernard Lidický, Nitya Mani and Yufei Zhao.
Apr. 16, 1:00pm, in WWH, Room 202 (this is a Tuesday!)

  • Venkatesan Guruswami (UC Berkeley/Simons Institute)
  • Abstract: The Parameterized Inapproximability Hypothesis (PIH) asserts that no fixed parameter tractable (FPT) algorithm can distinguish a satisfiable CSP instance, parameterized by the number of variables, from one where every assignment fails to satisfy 1% of the constraints. PIH plays the role of the PCP theorem in parameterized complexity. In this talk, we will sketch how to prove PIH under the Exponential Time Hypothesis (ETH). The approach involves two broad steps. We identify a highly structured ETH-hard CSP whose variables take vector values, and whose constraints are either parallel or linear. Both kinds of constraints are then checked with constant soundness via a "parallel PCP of proximity" based on the Hadamard code. Together these steps establish that ETH implies PIH. In a recent follow-up, we show that Reed-Muller code based PCPs can be devised to even get near-optimal runtime lower bounds, eg. showing that constant factor approximations to k-clique on n-vertex graphs requires n^{k^{1-o(1)}} time under ETH. Based on joint works with Bingkai Lin, Xuandi Ren, Yican Sun, Kewen Wu.
Apr. 18
    • Michal Koucký (Charles University)
    • Abstract: In this talk I will present a new locally consistent decomposition of strings with applications to sketching of edit distance and approximate pattern matching. Edit distance is a string similarity measure that counts how many characters have to be deleted, inserted or substituted in one string to get another one. Our ultimate goal is to be able to compute for each string a short sketch such that from sketches of two strings we can estimate their edit distance. Our new string decomposition allows to build such sketches of certain length.

      For a parameter k, each string x is decomposed into blocks that can be described by grammars of size O(k) (using some amount of randomness). If we take two strings x and y of edit distance at most k then their block decomposition uses the same number of grammars and the i-th grammar of x is the same as the i-th grammar of y except for at most k indexes i. The edit distance of x and y equals to the sum of edit distances of pairs of blocks where x and y differ. This decomposition can be used to build edit distance sketches and streaming approximate pattern matching algorithms.

      Joint work with Sudatta Bhattacharya.
Apr. 25
    • Van Vu (Yale University)
    • Abstract: Let \(A\) be a large matrix, and E is a random matrix. The aim is to compare \(f(A)\) and \(f(A+E)\) where \(f\) is a matrix functional. (For instance, one can take \(f(A)= A_k\), the best rank \(k\) approximation of \(A\).) We study the case when \(A\) has a rapidly decaying spectrum (a special case is \(A\) has low rank), which is quite common in practice, and obtain, for various functional \(f\), new bounds which significantly improve those obtained by traditional methods. We will discuss these results with applications, including the multi hidden cliques problem, PCA analysis for spiked sample covariance matrices, and private matrix approximation.
May 2
  • Rotem Oshman (Tel-Aviv University, Princeton)
  • Abstract: In distributed certification, our goal is to certify that a network has a certain desired property, e.g., the network is connected, or the internal states of its nodes encode a valid spanning tree of the network. To this end, a prover generates certificates that are stored at each network node, and the nodes can then interact with one another in order to check their certificates and verify that the property holds. This can be viewed as a distributed analog of NP. However, like most theoretical models of distributed computing, the traditional model for distributed certification is information-theoretic: the prover and the network nodes are computationally unbounded, and the only efficiency measures we consider are the length of the certificates and the communication required to verify them. The information-theoretic nature of the model sets the area of distributed certification apart from classical complexity theory, and also makes it subject to strong lower bounds from the world of nondeterministic communication complexity.

    In this talk I will describe a computationally sound version of distributed certification, where the prover and the network nodes are required to run in polynomial time in the size of the network. Introducing computational restrictions allows us to leverage cryptography, and we show that any network property in P can be certified using certificates of polylogarithmic length by a global prover that sees the entire network. Ideally, however, we would like the prover to also be distributed, and we show that this can be done as well: any polynomial distributed algorithm can efficiently certify its own execution with low overhead in terms of communication and rounds. The main challenge in the distributed setting is that no single network node can see the entire execution of the algorithm.

    This is joint work with Eden Aldema Tshuva, Elette Boyle, Ran Cohen and Tal Moran. No background in cryptography is assumed for the talk.
May 23
  • Sorrachai Yingchareonthawornchai (Hebrew University of Jerusalem)
  • Abstract: Vertex connectivity \(\kappa\) of an \(n\)-vertex \(m\)-edge undirected graph is the minimum number of vertices whose removal disconnects the graph. After more than half a century of intensive research, a randomized algorithm for computing vertex connectivity in near-optimal \(m^{1+o(1)}\) time was recently shown [LNPSY, STOC'21] via a (randomized) reduction to the max-flow problem. Deterministic algorithms, unfortunately, have remained much slower even though the max-flow problem can be solved deterministically in almost linear time [BCGKLPSS, FOCS'23]. When \(\kappa \le 2\), linear-time algorithms were known [Tarjan'72; Hopcroft Tarjan'73]. For \(\kappa \geq 3\), the state-of-the-art deterministic algorithms run either in \(m^{1+o(1)} \cdot 2^{O(\kappa^2)}\) time [Saranurak, Yingchareonthawornchai, FOCS'22] or in time proportional to \(\tilde O(n + \kappa\cdot \min\{\kappa, \sqrt{n}\})\) [Even'75; Gabow, FOCS'00] max-flow calls where we use \(\tilde O(\cdot)\) to hide a poly-logarithmic factor.

    In this talk, I will present a deterministic algorithm for computing vertex connectivity in time proportional to \(\kappa \cdot n^{o(1)}\) max-flow calls. For all values of \(\kappa\), our algorithm subsumes (up to a sub-polynomial factor) all known state-of-the-art algorithms [Tarjan'72; Hopcroft Tarjan'73; Even'75; Gabow, FOCS'00; Saranurak, Yingchareonthawornchai, FOCS'22]. In particular, our algorithm runs in \(m^{1+o(1)}\cdot n\) time in the worst-case, answering the open question asked by [Gabow, JACM'06].

    To obtain our result, we introduce a novel technique called common-neighborhood clustering and crucially use it for both reducing the number of max-flow calls, and reducing the size of max-flow instances per call. Furthermore, we show how to exploit Ramanujan expanders more efficiently than the previous technique of Gabow [FOCS'00] and introduce the pseudorandom object called selector into the context of vertex connectivity.

    Joint work with Yonggang Jiang, Chaitanya Nalam, and Thatchaphol Saranurak.
June 20 (summer special!)
Room 517, WWH
  • Nishant Mehta (University of Victoria)
  • Abstract: The sleeping bandits problem is a variant of the classical multi-armed bandit problem. In each of a sequence of rounds: an adversary selects a set of arms (actions) which are available (unavailable arms are "asleep"), the learning algorithm (Learner) then pulls an arm, and finally the adversary sets the loss of each available arm. Learner suffers the loss of the selected arm and observes only that arm's loss. A standard performance measure for this type of problem with changing action sets is the per-action regret: the learning algorithm wishes, for all arms simultaneously, to have cumulative loss not much larger than the cumulative loss of the arm when considering only those rounds in which the arm was available. For this problem, I'll present the first algorithms which enjoy low per-action regret against reactive (i.e., adaptive) adversaries; moreover, the regret guarantees are near-optimal, unlike previous results which were far from optimal even for oblivious adversaries. Along the way, we will review the simpler, full-information version of the problem, known as sleeping experts. I will also mention various extensions of our results, including (i) new results for bandits with side information from sleeping experts; (ii) guarantees for the adaptive regret and tracking regret for standard (non-sleeping) bandits. The results described so far are joint work with my PhD student Quan Nguyen (who worked out most of the results). Time permitting, I may start the talk by presenting an elegant lower bound for a related setting (dying experts, where experts only disappear over time) when using a different, much harder notion of regret known as the ranking regret.
Sept. 7
  • Zihan Tan (Rutgers University)
  • Abstract: Given a large graph \(G\) with a subset \(|T|=k\) of its vertices called terminals, a quality-\(q\) flow sparsifier is a small graph \(G'\) that contains \(T\) and preserves all multicommodity flows that can be routed between terminals in \(T\), to within factor \(q\). The problem of constructing flow sparsifiers with good (small) quality and (small) size has been a central problem in graph compression for decades.

    A natural approach of constructing \(O(1)\)-quality flow sparsifiers, which was adopted in most previous constructions, is contraction. Andoni, Krauthgamer, and Gupta constructed a sketch of size \(f(k,\epsilon)\) that stores all feasible multicommodity flows up to factor \((1+\epsilon)\), raised the question of constructing quality-\((1+\epsilon)\) flow sparsifiers whose size only depends on \(k\),\(\epsilon\) (but not the number of vertices in the input graph \(G\)), and proposed a contraction-based framework towards it using their sketch result. In this paper, we settle their question for contraction-based flow sparsifiers, by showing that quality-\((1+\epsilon)\) contraction-based flow sparsifiers with size \(f(\epsilon)\) exist for all 5-terminal graphs, but not all 6-terminal graphs. Our hardness result on 6-terminal graphs improves upon a recent hardness result by Krauthgamer and Mosenzon on exact (quality-1) flow sparsifiers, for contraction-based constructions. Our construction and proof utilize the notion of tight span in metric geometry.

    This talk is based on joint work with Yu Chen.
Sept. 14
  • Tal Yankovich (Tel Aviv University)
  • Abstract: Codes with locality play a major role in modern coding theory. Examples for such codes are locally decodable codes (LDC), locally correctable codes (LCC), locally testable codes (LTC), relaxed LDC and relaxed LCC. In this talk we will present a few results on codes with locality, for example, asymptotically good relaxed locally correctable codes with query complexity \(\(log n)^{2+o(1)}\).

    Based on joint works with Gil Cohen.
Sept. 21
  • Henry Yuen (Columbia University)
  • Abstract: How hard is it to compress quantum information? To produce a counterfeit quantum money state? To unscramble the Hawking radiation of a black hole? Traditional complexity theory -- which is centered around decision problems and tasks with classical inputs and outputs -- appears inadequate for reasoning about the complexity of such tasks involving quantum inputs and outputs.

    In this talk I will discuss why we need a "fully quantum" complexity theory, and will describe some facets of such a theory. As a key illustration I will explain how a "fully quantum" task called the Uhlmann Transformation Problem characterizes the complexity of seemingly unrelated problems, such as decoding noisy quantum channels, performing the Harlow-Hayden black hole radiation decoding task, and breaking the security of quantum bit commitment schemes. Along the way I will describe many open problems and directions to explore in the world of fully quantum complexity theory.

    No prior quantum background will be assumed. Joint work with John Bostanci, Yuval Efron, Tony Metger, Alexander Poremba, and Luowen Qian (https://arxiv.org/abs/2306.13073).
Sept. 28
  • Hanna Komlós (Rutgers University)
  • Abstract: The online list-labeling problem is a classical combinatorial problem with a large literature of upper bounds, lower bounds, and applications. The goal is to store a dynamically-changing set of n items in an array of m slots, while maintaining the invariant that the items appear in sorted order, and while minimizing the relabeling cost, defined to be the number of items that are moved per insertion/deletion. There has long existed a gap between the lower and upper bounds in the most algorithmically interesting part of the problem's parameter space. We present our results, which narrow this gap for the first time in nearly 4 decades. Additionally, we will describe some recent results on composing multiple list-labeling algorithms in order to obtain the best combination of their respective cost guarantees.
Oct. 5
  • Cosmin Pohoata (Emory University)
  • Abstract: We discuss a new upper bound for the Heilbronn triangle problem, showing that for sufficiently large $n$ in every configuration of \(n\) points chosen inside a unit square there exists a triangle of area less than \(n^{-8/7-1/2000}\). This is joint work with Alex Cohen and Dmitrii Zakharov.
Oct. 12
  • Liana Yepremyan (Emory University)
  • Abstract: A graph is Ramsey if its largest clique or independent set is of size logarithmic in the number of vertices. While almost all graphs are Ramsey, there is still no known explicit construction of Ramsey graphs. We discuss recent work on some questions in this area. Along the way, we study some fundamental problems in additive combinatorics, and discover that group structure is superfluous for these problems. Joint work with David Conlon, Jacob Fox and Huy Tuan Pham.
Oct. 19
  • Tamalika Mukherjee (Columbia University)
  • Abstract: We develop a framework for efficiently transforming certain approximation algorithms into differentially-private variants, in a black-box manner. Specifically, our results focus on algorithms \(A\) that output an approximation to a function \(f\) of the form \((1−a)f(x)−k\leq A(x)\leq(1+a)f(x)+k\), where \(k\in R_{\geq 0}\) denotes additive error and \(a\in[0,1)\) denotes multiplicative error can be "tuned" to small-enough values while incurring only a polynomial blowup in the running time/space. We show that such algorithms can be made DP without sacrificing accuracy, as long as the function f has small global sensitivity. We achieve these results by applying the smooth sensitivity framework developed by Nissim, Raskhodnikova, and Smith (STOC 2007).

    Our framework naturally applies to transform non-private FPRAS and FPTAS algorithms into \(\epsilon\)-DP approximation algorithms where the former case requires an additional postprocessing step. We apply our framework in the context of sublinear-time and sublinear-space algorithms, while preserving the nature of the algorithm in meaningful ranges of the parameters. Our results include the first (to the best of our knowledge) \(\epsilon\)-edge DP sublinear-time algorithm for estimating the number of triangles, the number of connected components, and the weight of a minimum spanning tree of a graph. In the area of streaming algorithms, our results include ϵ-DP algorithms for estimating Lp-norms, distinct elements, and weighted minimum spanning tree for both insertion-only and turnstile streams. Our transformation also provides a private version of the smooth histogram framework, which is commonly used for converting streaming algorithms into sliding window variants, and achieves a multiplicative approximation to many problems, such as estimating Lp-norms, distinct elements, and the length of the longest increasing subsequence.
Oct. 26
  • Yotam Gafni (Technion)
  • Abstract: Decentralized cryptocurrencies rely on aligning the incentives of users and miners to operate correctly and offer a high QoS to users. Recent literature studies the mechanism design problem of the auction serving as a cryptocurrency's Transaction Fee Mechanism (TFM). We find that the non-myopic modeling of miners falls close to the well-known problem of online buffer management for packet switching. The main difference is that unlike packets, which are of a fixed size throughout their lifetime, in a financial environment user preferences (and therefore revenue extraction) may be time-dependent. We study the competitive ratio guarantees given a certain discount rate and show how existing methods from packet scheduling perform sub-optimally in the more general discounted setting. We find a novel, simple, memoryless, and competitively optimal deterministic algorithm for the semi-myopic case, as well as a randomized algorithm that achieves better performance than the best possible deterministic algorithm, for any discount rate
Nov. 2
  • Edinah Gnang (Johns Hopkins University)
  • Abstract: We describe proof of the long standing Kotzig–Ringel–Rosa conjecture (1964) as well the slightly more recent Graham-Sloane Conjectures (1980) respectively known as the graceful tree labeling conjecture and the harmonious tree labeling conjecture. Both proof stem from a functional reformulation of these conjectures and a marriage of Noga Alon's Combinatorial Nullstellensatz with a new composition lemma. If time permits we will also discuss algorithmic aspects of the composition lemma.
Nov. 9
  • Michael Simkin (Massachusetts Institute of Technology)
  • Abstract: The \(n\)-queens problem is to determine \(Q(n)\), the number of ways to place \(n\) mutually non-threatening queens on an \(n\) x \(n\) chess board. We show that there exists a constant \(1.94 < a < 1.9449\) such that \(Q(n) = ((1 + o(1))ne^(-a))^n\). The constant \(a\) is characterized as the solution to a convex optimization problem in \(P([-1/2,1/2]^2)\), the space of Borel probability measures on the square. The chief innovation is the introduction of limit objects for \(n\)-queens configurations, which we call "queenons". These are a convex set in \(P([-1/2,1/2]^2)\). We define an entropy function that counts the number of \(n\)-queens configurations approximating a given queenon. The upper bound uses the entropy method of Radhakrishnan and Linial-Luria. For the lower bound we describe a randomized algorithm that constructs a configuration near a prespecified queenon and whose entropy matches that found in the upper bound. The enumeration of \(n\)-queens configurations is then obtained by maximizing the (concave) entropy function over the space of queenons.
Nov. 16
  • Greg Rosenthal (University of Warwick, University of Cambridge)
  • Abstract: We present a polynomial-time quantum algorithm making a single query (in superposition) to a classical oracle, such that for every state \(|\psi \rangle\), there exists a choice of oracle that makes the algorithm construct an exponentially close approximation of \(|\psi \rangle\). Previous algorithms for this problem either used a linear number of queries and polynomial time, or a constant number of queries and polynomially many ancillae but no nontrivial bound on the runtime. As corollaries we do the following:
    • We simplify the proof that \(statePSPACE \subseteq stateQIP\) (a quantum state analogue of \(PSPACE \subseteq IP\)) and show that a constant number of rounds of interaction suffices.
    • We show that QACf0 lower bounds for constructing explicit states would imply breakthrough circuit lower bounds for computing explicit boolean functions.
    • We prove that every \(n\)-qubit state can be constructed to within 0.01 error by an \(O(2^n/n)\)-size circuit over an appropriate finite gate set. More generally we give a size-error tradeoff which, by a counting argument, is optimal for any finite gate set.
    To appear in SODA 2024. Available at https://arxiv.org/abs/2306.01723.
Nov. 23
  • No Meeting -- Thanksgiving.
Nov. 30
  • Victor Reis (Institute for Advanced Study)
  • Abstract: We prove that there exists an online algorithm that for any sequence of vectors \(v_1, ..., v_T\) in \(\mathbb{R}^n\) of Euclidean norm at most 1, arriving one at a time, decides random signs \(x_1,..., x_T \in \{−1,1\}\) so that for every \(t \leq T\), their signed prefix sum is 10-subgaussian. This improves over the work of Alweiss, Liu and Sawhney who kept prefix sums \(O(\sqrt{\log(nT)})\)-subgaussian, and gives a \(O(\sqrt{\log T})\) bound on the discrepancy \(max_{t \le T} \|\sum_{i=1}^t x_i v_i\|_\infty\). Our proof combines a generalization of Banaszczyk’s prefix balancing result to trees with a cloning argument to find distributions rather than single colorings. We also show a matching \(\Omega(\sqrt{\log T})\) strategy for an oblivious adversary.
Dec. 7
  • Kevin Pratt (New York Univeristy)
  • Abstract: In this talk I will give a brief overview of the area for fast matrix multiplication, with an emphasis on connections to additive combinatorics. After discussing connections between the current main approach to fast matrix multiplication and the resolution of the cap-set problem, I will raise the following question: suppose that \(S \subseteq [n]^2\) contains no three points of the form \((x,y), (x,y+\delta), (x+\delta,y')\), where \(\delta \neq 0\). How big can \(S\) be? Trivially, \(n \le |S| \le n^2\). Slight improvements on these bounds are obtained from Shkredov's upper bound for the corners problem, which shows that \(|S| \le O(n^2/(\log \log n)^c)\) for some small \(c > 0\), and a construction due to Petrov, which shows that \(|S| \ge \Omega(n \log n/\sqrt{\log \log n})\). Could it be that for all \(\varepsilon > 0\), \(|S| \le O(n^{1+\varepsilon})\)? I will sketch that if so, this would rule out obtaining \(\omega = 2\) via an approach of Cohn, Umans, Kleinberg, and Szegedy.
Dec. 14
  • Matija Bucic (Princeton University)
  • Abstract: Expander graphs are perhaps one of the most widely useful classes of graphs ever considered. In this talk, we will focus on a fairly weak notion of expanders called sublinear expanders, first introduced by Komlós and Szemerédi around 30 years ago. They have found many remarkable applications ever since. In particular, we will focus on certain robustness conditions one may impose on sublinear expanders and some applications of this idea, which include:
    • recent progress on the classical Erdős-Gallai cycle decomposition conjecture
    • an essentially tight answer to the classical Erdős unit distance problem for "most" real normed spaces and
    • an asymptotic solution to the rainbow Turan problem for cycles, raised by Keevash, Mubayi, Sudakov and Verstraete, with an interesting corollary in additive number theory.
Feb. 2
  • Roie Levin (Tel Aviv University)
  • Abstract: We give a polynomial-time algorithm for online covering IPs with a competitive ratio of \(O(\log mn)\) when the constraints are revealed in random order, essentially matching the best possible offline bound of \(O(\log n)\) and circumventing the \(\Omega(\log m \log n)\) lower bound known in adversarial order. We then use this result to give an \(O(log mn)\) competitive algorithm for the prophet version of this problem, where constraints are sampled from a sequence of known distributions (in fact, our algorithm works even when only a single sample from each of the distributions is given). Since our algorithm is universal, as a byproduct we establish that only \(O(n)\) samples are necessary to build a universal map for online covering IPs with competitive ratio O(\log mn) on input sequences of length n.

    This talk is based on joint work with Anupam Gupta and Gregory Kehne, the first half of which appeared at FOCS 2021.
Feb. 9
  • Nati Linial (Hebrew University of Jerusalem)
  • Abstract: The underlying idea of this lecture is that all important phenomena in geometry have interesting graph-theoretic counterparts. A path system in an n-vertex graph \(G=(V,E)\) is a collection of \({n \choose 2}\) paths \(Puv=Pvu\), one per each pair of vertices \(u,v\), where \(Puv\) connects the vertices \(u\) and \(v\). We say that \(P\) is consistent if it is closed under taking subpaths. Namely, for any vertex \(x\) that resides on \(Puv\), it holds that \(Puv\) is the concatenation of \(Pux\) and \(Pxv\). There is a very simple way to generate consistent path systems. Namely, pick a positive weight function \(w\) on the edge set \(E\), and let \(Puv\) be a \(w\)-shortest \(uv\) path. A path system that can be attained this way, is said to be metric. Question: Are all consistent path systems metric? Answer: A very emphatic NO. We call \(G\) metrizable if every consistent path system in \(G\) is metric. Our main discoveries are:
    1. 1. Almost all graphs (at a very strong sense) are non-metrizable.
    2. Yet, all outerplanar graphs are metrizable.
    3. Metrizability is polynomial-time decidable.
    This is joint work with Daniel Cizma
Feb. 16
  • Prantar Ghosh (Rutgers University)
  • Abstract: A streaming algorithm is called adversarially robust if it provides correct outputs even when the stream updates are chosen by an adaptive adversary based on past outputs given by the algorithm. There has been a surge in interest for such algorithms over the last couple of years. I shall begin this talk by describing this model and then address the natural question of exhibiting a separation between classical and robust streaming. We shall see how this question leads to the problem of streaming graph coloring, where we need to maintain a proper vertex-coloring of a streaming graph using as few colors as possible. In the classical streaming model, Assadi, Chen, and Khanna showed that an n-vertex graph with maximum degree Δ can be colored with Δ+1 colors in O(n polylog(n)) space, i.e., semi-streaming space. We shall show that an adversarially robust algorithm running under a similar space bound must spend almost Ω(Δ^2) colors, and that robust O(Δ)-coloring requires a linear amount of space, namely Ω(nΔ). These lower bounds not only establish the first separation between adversarially robust algorithms and ordinary randomized algorithms for a natural problem on insertion-only streams, but also the first separation between randomized and deterministic coloring algorithms for graph streams: this is because deterministic algorithms are automatically robust. I shall also go over some complementary upper bounds: in particular, there are robust coloring algorithms using O(Δ^2.5) colors in semi-streaming space and O(Δ^2) colors in O(n√Δ) space.

    This is based on a couple of joint works: one with Amit Chakrabarti and Manuel Stoeckl, and another with Sepehr Assadi, Amit, and Manuel.
Feb. 23
  • Spencer Peters (Cornell University)
  • Abstract: Function inversion is a fundamental problem in cryptography and in theoretical computer science more broadly. Given a function f: [N] -> [N] from a finite domain to itself, the goal is to construct a small data structure, so that for any point y in the image of f, you can recover an inverse of y using few evaluations of f. First, I will describe a modification to Fiat and Naor's classic function inversion algorithm [STOC '91] that improves the time-space tradeoff in the regime where the number of evaluations T exceeds the data structure bitlength S. Then I'll present the first (barely) non-trivial non-adaptive algorithm for function inversion (a non-adaptive algorithm chooses all the points it will evaluate f on before seeing any of the results). This algorithm resolves a question posed by Corrigan-Gibbs and Kogan [TCC'19], and I'll show that the tradeoff it achieves is tight for a natural class of non-adaptive algorithms. Both results are joint work with Alexander Golovnev, Siyao Guo, and Noah Stephens-Davidowitz.
Mar. 2
  • Mark Braverman (Princeton University)
  • Abstract: Our goal is to develop a generic framework for converting modern gradient-descent based optimization algorithms into mechanisms where inputs come from self-interested agents. We focus on aggregating preferences from n players in a context without money. Special cases of this setting include voting, allocation of items by lottery, and matching. Our key technical contribution is a new meta-algorithm we call APEX (Adaptive Pricing Equalizing Externalities). The framework is sufficiently general to be combined with any optimization algorithm that is based on local search. In the talk I'll outline the algorithm, and open problem/research directions that it raises. As an application of the framework, I will discuss a special case of applying the framework to the problem of one-sided allocation with lotteries. In this case, we obtain a strengthening of the 1979 result by Hylland and Zeckhauser on allocation via a competitive equilibrium from equal incomes (CEEI). The [HZ79] result posits that there is a (fractional) allocation and a set of item prices such that the allocation is a competitive equilibrium given prices. We further show that there is always a reweighing of the players' utility values such that running the standard unit-demand VCG with reweighed utilities leads to a HZ-equilibrium prices. Interestingly, not all HZ competitive equilibria come from VCG prices.
Mar. 9
  • No Meeting
Mar. 23
  • Karthik C.S. (Rutgers University)
  • Abstract: In the Discrete Steiner Tree problem (DST), we are given as input two sets of points in a metric space, called terminals and facilities respectively, and the goal is to find the minimum-cost tree connecting the terminals, by possibly introducing new points (called Steiner points) from the set of facilities, as nodes in the solution. In the Continuous Steiner Tree problem (CST) we are only given as input a set of terminals and are free to introduce Steiner points in the solution tree from anywhere in the metric space.

    Determining the efficiency of (even approximately) computing Steiner trees has received a lot of attention from the theoretical computer science community. Despite this focus, our understanding of the approximability of Steiner trees remains poor, and our understanding of the hardness of approximation of these tasks was even worse. In this talk, I will detail some recent progress on the hardness of approximating DST and CST when the input points are in \(L_p\) metric spaces.

    The talk is based on joint work with Henry Fleischmann and Surya Teja Gavva.
Mar. 30
  • Or Zamir (Institute for Advanced Study)
  • Abstract: We present a general method to convert algorithms into faster algorithms for almost-regular input instances. Informally, an almost-regular input is an input in which the maximum degree is larger than the average degree by at most a constant factor. This family of inputs vastly generalizes several families of inputs for which we commonly have improved algorithms, including bounded-degree inputs and random inputs. It also generalizes families of inputs for which we don't usually have faster algorithms, including regular-inputs of arbitrarily high degree and all very dense inputs. We apply our method to achieve breakthroughs in exact algorithms for several central NP-Complete problems including k-SAT, Graph Coloring, and Maximum Independent Set. Our main tool is the first algorithmic application of the relatively new Hypergraph Container Method (Saxton and Thomason 2015, Balogh, Morris and Samotij 2015). This recent breakthrough, which generalizes an earlier version for graphs (Kleitman and Winston 1982, Sapozhenko 2001), has been used extensively in recent years in extremal combinatorics. An important component of our work is a new generalization of (hyper-)graph containers to Partition Containers.
Apr. 6
  • Gil Cohen (Tel Aviv Univeristy)
  • Abstract: Random walks on expanders are extremely useful in theoretical computer science, and beyond. Unfortunately though, they have an inherent cost. E.g., the spectral expansion of a Ramanujan graph deteriorates exponentially with the length of the walk. In this talk we will see how this exponential cost can be reduced to linear by applying a permutation after each random step. These permutations are tailor-made to the graph at hand, requiring no randomness to generate.

    Our proof is established using the powerful framework of finite free probability and interlacing families that was introduced by Marcus, Spielman, and Srivastava in their seminal sequence of works on bipartite Ramanujan graphs.

    Joint work with Gal Maor. Link to paper: https://eccc.weizmann.ac.il/report/2022/163/.
Apr. 13
  • Misha Khodak (Carnegie Mellon University)
  • Abstract: A burgeoning paradigm in algorithm design is learning-augmented algorithms, or algorithms with predictions, where methods can take advantage of a (possibly imperfect) prediction about their instance. While past work has focused on using predictions to improve competitive ratios and runtime, this talk addresses a different, salient question: how do we learn the predictions themselves? We introduce an approach for co-designing learning-augmented algorithms with their own custom learning algorithms, with the crucial step being to optimize nice surrogate losses bounding the algorithms’ costs. This leads to improved sample complexity bounds for several learning-augmented graph algorithms and the first learning-theoretic guarantees for page migration with predictions, among other contributions. We also instantiate these ideas on the new direction of learning-augmented private algorithms, where the goal is to reduce utility loss due to privacy rather than runtime. Our approach drives numerous insights on how to robustly incorporate external information to release better statistics of sensitive datasets, which we verify empirically on the task of multiple quantile release.
Apr. 19, 3:00pm, in 60 5th Avenue, Room C15 (this is a Wednesday)

  • Rachel Cummings (Columbia University)
  • Abstract: The problem of learning from imbalanced datasets, where the label classes are not equally represented, arises often in practice. A widely used approach to address this problem is to rely on pre-processing techniques to amplify the minority class, e.g., by reusing or resampling the existing minority instances. However, when the training data are sensitive and privacy is a concern, this will also amplify privacy leakage from members of the minority class. Therefore, oversampling pre-processing steps must be designed with downstream privacy in mind. Our work provides insights about the implications of pre-processing before the use of DP algorithms, which we hope will be of broad interest.

    In this paper, we first quantify the privacy degradation from running a commonly used Synthetic Minority Oversampling TEchnique (SMOTE) as a pre-processing step before applying any differentially private algorithm. We show that this degregation increases exponentially in the dimensionality of the data, which harms the accuracy of downstream private learning at a corresponding rate. We then present a differentially private variant of this algorithm (DP-SMOTE) that guarantees formal privacy for the minority class, even during oversampling. Finally, we show empirically that, when applied as a pre-processing step to highly imbalanced data before private regression, DP-SMOTE outperformed SMOTE and other baseline methods, due primarily to the privacy degradation in the non-private pre-processing methods.

    joint work with Yuliia Lut, Ethan Turok, and Marco Avella-Medina
Apr. 27
  • No Meeting
May 4
  • Eric Balkanski (Columbia University)
  • Abstract: We introduce an alternative model for the design and analysis of strategyproof mechanisms that is motivated by the recent surge of work in "learning-augmented algorithms". Aiming to complement the traditional approach in computer science, which analyzes the performance of algorithms based on worst-case instances, this line of work has focused on the design and analysis of algorithms that are enhanced with machine-learned predictions regarding the optimal solution. The algorithms can use the predictions as a guide to inform their decisions, and the goal is to achieve much stronger performance guarantees when these predictions are accurate (consistency), while also maintaining near-optimal worst-case guarantees, even if these predictions are very inaccurate (robustness). So far, these results have been limited to algorithms.

    We initiate the design and analysis of strategyproof mechanisms that are augmented with predictions regarding the private information of the participating agents. To exhibit the important benefits of this approach, we revisit the canonical problems of strategic facility location and strategic scheduling. We propose new strategyproof mechanisms that leverage predictions to guarantee an optimal trade-off between consistency and robustness guarantees. Furthermore, we also prove parameterized approximation results as a function of the prediction error, showing that our mechanisms perform well even when the predictions are not fully accurate.

    Joint work with Priyank Agrawal, Vasilis Gkatzelis, Tingting Ou, and Xizhi Tan
May 11
  • Grigory Yaroslavtsev (George Mason University)
  • Abstract: How many labeled tuples are required for learning a good representation of the data? This question arises in similarity learning via contrastive sampling, in algorithms for learning tree representations, etc. I will present tight bounds on the number of samples required for these problems, including learning general metric spaces, Euclidean and lp-spaces, tree metrics and other tree representations.

    Joint work with Dmitrii Avdiukhin, Sauman Das, Orr Fischer, Faraz Mirza and Danny Vainstein.
Sept. 22
  • Vladimir Podolskii (NYU)
  • Abstract: We consider sorting networks that are constructed from comparators of arity k>2. That is, in our setting the arity of the comparators — or, in other words, the number of inputs that can be sorted at the unit cost — is a parameter. We study its relationship with two other parameters — n, the number of inputs, and d, the depth. This model received considerable attention. Partly, its motivation is to better understand the structure of sorting networks. In particular, sorting networks with large arity are related to recursive constructions of ordinary sorting networks. Additionally, studies of this model have natural correspondence with a recent line of work on constructing circuits for majority functions from majority gates of lower fan-in.

    We obtain the first lower bounds on the arity of constant-depth sorting networks. More precisely, we consider sorting networks of depth d up to 4, and determine the minimal k for which there is such a network with comparators of arity k. For depths d=1, 2 we observe that k=n. For d=3 we show that k=n/2. For d=4 the minimal arity becomes sublinear: k=\Theta(n^{2/3}). This contrasts with the case of majority circuits, in which k=O(n^{2/3}) is achievable already for depth d=3.

    Joint work with Natalia Dobrokhotova-Maikova and Alexander Kozachinskiy: https://eccc.weizmann.ac.il/report/2022/116/

    Bio: Vladimir Podolskii defended his PhD thesis in 2009 in Moscow State University. His research areas are computational complexity, tropical algebra, applications of complexity theory to databases augmented with logical theories. Until recently he worked in Steklov Mathematical Institute (Moscow) and HSE University (Moscow).
Sept. 29
  • Sepehr Assadi (Rutgers)
  • Abstract: Recent breakthroughs in graph streaming have led to the design of single-pass semi-streaming algorithms for various graph coloring problems such as (Δ+1)-coloring, Δ-coloring, degeneracy-coloring, coloring triangle-free graphs, and others. These algorithms are all randomized in crucial ways and whether or not there is any deterministic analogue of them has remained an important open question in this line of work.

    In this talk, we will discuss our recent result that fully resolves this question: there is no deterministic single-pass semi-streaming algorithm that given a graph G with maximum degree Δ, can output a proper coloring of G using any number of colors which is sub-exponential in Δ. The proof is based on analyzing the multi-party communication complexity of a related communication game using elementary random graph theory arguments.

    Based on joint work with Andrew Chen (Cornell) and Glenn Sun (UCLA): https://arxiv.org/abs/2109.14891A.
Oct. 6
  • Peng Zhang (Rutgers)
  • Abstract: Marcus, Spielman, and Srivastava (Annals of Mathematics, 2015) solved the Kadison-Singer Problem by proving a strong form of Weaver’s discrepancy conjecture. They showed that for all \(\alpha > 0\) and all lists of vectors of norm at most \(\sqrt{\alpha}\) whose outer products sum to the identity, there exists a signed sum of those outer products with operator norm at most \(\sqrt{8 \alpha} + 2 \alpha\). Besides its relation to the Kadison-Singer problem, Weaver’s discrepancy problem has applications in graph sparsification and randomized experimental design. In this talk, we will prove that it is NP-hard to distinguish such a list of vectors for which there is a signed sum that equals the zero matrix from those in which every signed sum has operator norm at least \(k \sqrt{\alpha}\), for some absolute constant \(k > 0\). Thus, it is NP-hard to construct a signing that is a constant factor better than that guaranteed to exist. This result is joint work with Daniel Spielman.
Oct. 13
  • Dominik Kempa (Stony Brook)
  • Abstract: Lempel-Ziv (LZ77) compression is one of the most commonly used lossless compression algorithms. The basic idea is to greedily break the input string into blocks (called phrases), every time forming as a phrase the longest prefix of the unprocessed part that has an earlier occurrence. In 2010, Kreft and Navarro introduced a variant of LZ77 called LZ-End, which requires the previous occurrence of each phrase to end at the boundary of an already existing phrase. They conjectured that it achieves a compression that is always close to LZ77. In this talk, we: (1) present the first proof of this conjecture; (2) discuss the first data structure implementing fast random access to the original string using space linear in the size of LZ-End parsing. We will also give a broad overview of the increasingly popular field of compressed data structures/algorithms.

    This is joint work with Barna Saha (UC San Diego) and was presented at SODA 2022.
Oct. 18
(this is a Tuesday!)
Room 202.

  • Yuval Rabani (Hebrew University of Jerusalem)
  • Abstract: We prove new bounds, mostly lower bounds, on the competitive ratio of a few online problems, including the \(k\)-server problem and other related problems. In particular:

    1. The randomized competitive ratio of the \(k\)-server problem is at least \(\Omega(\log^2 k)\) in some metric spaces. This refutes the long-standing randomized \(k\)-server conjecture that the competitive ratio is \(O(\log k)\) in all metric spaces.

    2. Consequently, there is a lower bound of \(\Omega(\log^2 n)\) on the competitive ratio of metrical task systems in some \(n\)-point metric spaces, refuting an analogous conjecture that in all \(n\)-point metric spaces the competitive ratio is \(O(\log n)\). This lower bound matches asymptotically the best previously known universal upper bound.

    3. The randomized competitive ratio of traversing width-\(w\) layered graphs is \(\Theta(w^2)\). The lower bound improves slightly the previously best lower bound. The upper bound improves substantially the previously best upper bound.

    4. The \(k\)-server lower bounds imply improved lower bounds on the competitive ratio of distributed paging and metric allocation.

    5. The universal lower bound on the randomized competitive ratio of the \(k\)-server problem is \(\Omega(\log k)\). Consequently, the universal lower bound for \(n\)-point metrical task systems is \(\Omega(\log n)\). These bounds improve the previously known universal lower bounds, and they match asymptotically existential upper bounds.

    The talk is based on two papers which are both joint work with Sebastien Bubeck and Christian Coester: Shortest paths without a map, but with an entropic regularizer (the upper bound in 3, to appear in FOCS 2022) The randomized \(k\)-server conjecture is false! (all the lower bounds, manuscript)
Oct. 27
  • Jessica Sorrell (University of Pennsylvania)
  • Abstract: Reproducibility is vital to ensuring scientific conclusions are reliable, but failures of reproducibility have been a major issue in nearly all scientific areas of study in recent decades. A key issue underlying the reproducibility crisis is the explosion of methods for data generation, screening, testing, and analysis, where, crucially, only the combinations producing the most significant results are reported. Such practices (also known as p-hacking, data dredging, and researcher degrees of freedom) can lead to erroneous findings that appear to be significant, but that don’t hold up when other researchers attempt to replicate them.

    In this talk, we introduce a new notion of reproducibility for randomized algorithms. This notion ensures that with high probability, an algorithm returns exactly the same output when run with two samples from the same distribution. Despite the exceedingly strong demand of reproducibility, there are efficient reproducible algorithms for several fundamental problems in statistics and learning, including statistical queries, approximate heavy-hitters, medians, and halfspaces. We also discuss connections to other well-studied notions of algorithmic stability, such as differential privacy.

    This talk is based on prior and ongoing work with Mark Bun, Marco Gaboardi, Max Hopkins, Russell Impagliazzo, Rex Lei, Toniann Pitassi, and Satchit Sivakumar.
Nov. 3
  • Riko Jacob (IT University of Copenhagen)
  • Authors: Michael T. Goodrich, Riko Jacob, and Nodari Sitchinava

    Abstract: We prove an \(\Omega(\log n \log \log n)\) lower bound for the span of implementing the \(n\) input, \(\log n\)-depth FFT circuit (also known as butterfly network) in the nonatomic binary fork-join model. In this model, memory-access synchronizations occur only through fork operations, which spawn two child threads, and join operations, which resume a parent thread when its child threads terminate. Our bound is asymptotically tight for the nonatomic binary fork-join model, which has been of interest of late, due to its conceptual elegance and ability to capture asynchrony.Our bound implies super-logarithmic lower bound in the nonatomic binary fork-join model for implementing the butterfly merging networks used, e.g., in Batcher's bitonic and odd-even mergesort networks. This lower bound also implies an asymptotic separation result for the atomic and nonatomic versions of the fork-join model, since, as we point out, FFT circuits can be implemented in the atomic binary fork-join model with span equal to their circuit depth.
Nov. 10
  • Huacheng Yu (Princeton University)
  • Abstract: In this talk, we show a strong XOR lemma for bounded-round two-player randomized communication. For a function \(f:X\times Y\rightarrow\{0,1\}\), the n-fold XOR function \(f^{\oplus n}:X^n\times Y^n \rightarrow\{0,1\}\) maps n input pairs \((x_1,...,x_n), (y_1,...,y_n)\) to the XOR of the n output bits \(f(x_1,y_1)\oplus \ldots \oplus f(x_n, y_n)\). We prove that if every r-round communication protocols that computes f with probability 2/3 uses at least C bits of communication, then any r-round protocol that computes \(f^{\oplus n}\) with probability \(1/2+exp(-O(n))\) must use \(n(r^{-O (r)}C-1)\) bits. When r is a constant and C is sufficiently large, this is \(\Omega(nC)\) bits. It matches the communication cost and the success probability of the trivial protocol that computes the n bits \(f(x_i,y_i)\) independently and outputs their XOR, up to a constant factor in n. A similar XOR lemma has been proved for f whose communication lower bound can be obtained via bounding the discrepancy [Shaltiel03]. By the equivalence between the discrepancy and the correlation with 2-bit communication protocols, our new XOR lemma implies the previous result.
Nov. 17
  • Sophie Huiberts (Columbia University)
  • Abstract: Explaining why the simplex method is fast in practice, despite it taking exponential time in the theoretical worst case, continues to be a challenge. Smoothed analysis is a paradigm for addressing this question. During my talk I will present an improved upper bound on the smoothed complexity of the simplex method, as well as prove the first non-trivial lower bound on the smoothed complexity. This is joint work with Yin Tat Lee and Xinzhi Zhang.
Dec. 1
  • Roei Tell (Institute for Advanced Study/DIMACS)
  • Abstract: In the first half of the talk I'll set up some background, by describing two recent directions in the study of derandomization: A non-black-box algorithmic framework, which replaces the classical PRG-based paradigm; and free lunch results, which eliminate randomness with essentially no runtime overhead.

    In the second half we'll see one result along these directions: Under hardness assumptions, every doubly efficient proof system with constantly many rounds of interaction can be simulated by a deterministic NP-type verifier, with essentially no runtime overhead, such that no efficient adversary can mislead the deterministic verifier. Consequences include an NP-type verifier of this type for #SAT, running in time \(2^{\epsilon n}\) for an arbitrarily small constant eps>0; and a complexity-theoretic analysis of the Fiat-Shamir heuristic in cryptography.

    The talk is based on a joint work with Lijie Chen (UC Berkeley).
Dec. 8
  • Michael Chapman (NYU)
  • Abstract: A property is testable if there exists a probabilistic test that distinguishes between objects satisfying this property and objects that are far away from satisfying the property. In other words, if an object passes the test with high probability, it is close to an object that satisfies the test with probability 1 (namely, an object with the desired property). Property testing results are useful for many TCS applications, mainly in error correction, probabilistically checkable proofs and hardness of approximation.

    In this talk we are going to discuss two property testing problems that arise naturally in group theory. (All relevant group theoretic notions will be defined during the talk).

    1. The first is due to Ulam, who in 1940 asked: Given an almost homomorphism between two groups, is it close to an actual homomorphism between the groups? The answer depends on the choice of groups, as well as what we mean by "almost" and "close". We will present some classical and recent results in TCS in this framework, specifically the BLR test and quantum soundness of 2-player games. We will discuss some recent developments and open problems in this field.

    2. The second problem is the following: Is being a proper subgroup a testable property? We will discuss partial results and a very nice open problem in this direction.
Dec. 15
  • Aaron Sidford (Stanford)
  • Abstract: In this talk I will discuss recent advances in the fundamental robust optimization problem of minimizing the maximum of a finite number of convex loss functions. In particular I will show how to develop stochastic methods for approximately solving this problem with a near-optimal number of gradient queries. Along the way, I will cover several broader tools in the design and analysis of efficient optimization algorithms, including accelerated methods for using ball-optimization oracles and stochastic bias-reduced gradient methods. This talk will include joint work with Hilal Asi, Yair Carmon, Arun Jambulapati, and Yujia Jin including https://arxiv.org/abs/2105.01778 and https://arxiv.org/abs/2106.09481.
May 21
  • Jeroen Zuiddam (IAS)
  • Abstract: We give a dual description of the Shannon capacity of graphs. The Shannon capacity of a graph is the rate of growth of the independence number under taking the strong power, or in different language, it is the maximum rate at which information can be transmitted over a noisy communication channel. Our dual description gives Shannon capacity as a minimization over the "asymptotic spectrum of graphs", which as a consequence unifies previous results and naturally gives rise to new questions. Besides a gentle introduction to the asymptotic spectrum of graphs we will discuss, if time permits, Strassen's general theory of "asymptotic spectra" and the asymptotic spectrum of tensors.
May 16
  • Omri Weinstein (Columbia University)
  • Abstract: I will talk about new connections between arithmetic data structures, circuit lower bounds and pseudorandomness. As the main result, we show that static data structure lower bounds in the group (linear) model imply semi-explicit lower bounds on matrix rigidity. In particular, we prove that an explicit lower bound of t ω(log^2 n) on the cell-probe complexity of linear data structures in the group model, even against arbitrarily small linear space (s = (1+ε)n), would already imply a semi-explicit (P^NP) construction of rigid matrices with significantly better parameters than the current state of art (Alon, Panigrahy, and Yekhanin, 2009). Our result further asserts that polynomial (t n^δ) data structure lower bounds against near-maximal space, would imply super-linear circuit lower bounds for log-depth linear circuits (which would close a four-decade open question). In the succinct space regime (s = n+o(n)), we show that any improvement on current cell-probe lower bounds in the linear model would also imply new rigidity bounds. Our main result relies on a new connection between the inner and outer dimensions of a matrix (Paturi and Pudlak, 2006), and on a new worst-to-average case reduction for rigidity, which is of independent interest.

    Based mostly on joint work with Zeev Dvir (Princeton) and Alexander Golovnev (Harvard).
May 2
  • Noah Stephens-Davidowitz (MIT)
  • Abstract: We show that, assuming a common conjecture in complexity theory, there are "no non-trivial algorithms" for the two most important problems in coding theory: the Nearest Codeword Problem (NCP) and the Minimum Distance Problem (MDP). Specifically, for any constant \eps > 0, there is no N^{1-\eps}-time algorithm for codes with N codewords. In fact, the NCP result even holds for a family of codes with a single code of each cardinality, and our hardness result therefore also applies to the preprocessing variant of the problem.

    These results are inspired by earlier work showing similar results for the analogous lattice problems (joint works with three other NYU alums: Huck Bennett and Sasha Golovnev and with Divesh Aggarwal), but the proofs for coding problems are far simpler. As in those works, we also prove weaker hardness results for approximate versions of these problems (showing that there is no N^{o(1)}-time algorithm in this case).

    Based on joint work with Vinod Vaikuntanathan.
Apr. 11
  • Li-Yang Tan (Stanford University)
  • Abstract: We prove a new Littlewood--Offord-type anticoncentration inequality for m-facet polytopes, a high-dimensional generalization of the classic Littlewood--Offord theorem. Our proof draws on and extends techniques from Kane's bound on the boolean average sensitivity of m-facet polytopes. Joint work with Ryan O'Donnell and Rocco Servedio; manuscript available at https://arxiv.org/abs/1808.04035.
Mar. 21
  • Sahil Singla (Princeton)
  • Abstract: In the classical secretary problem, a sequence of n elements arrive in a uniformly random order. The goal is to maximize the probability of selecting the largest element (or to maximize the expected value of the selected item). This model captures applications like online auctions, where we want to select the highest bidder. In many such applications, however, one may expect a few outlier arrivals that are adversarially placed in the arrival sequence. Can we still select a large element with good probability? Dynkin’s popular 1/e-secretary algorithm is sensitive to even a single adversarial arrival: if the adversary gives one large bid at the beginning of the stream, the algorithm does not select any element at all. In this work we introduce the Byzantine Secretary problem where we have two kinds of elements: green (good) and red (bad). The green elements arrive uniformly at random. The reds arrive adversarially. The goal is to find a large green element. It is easy to see that selecting the largest green element is not possible even when a small fraction of the arrival is red, i.e., we cannot do much better than random guessing. Hence we introduce the second-max benchmark, where the goal is to select the second-largest green element or better. This dramatically improves our results. We study both the probability-maximization and the value-maximization settings. For probability-maximization, we show the existence of a good randomized algorithm, using the minimax principle. Specifically, we give an algorithm for the known distribution case, based on trying to guess the second-max in hindsight, and using this estimate as a good guess for the future. For value-maximization, we give an explicit poly log^∗ n competitive algorithm, using a multi-layered bucketing scheme that adaptively refines our estimates of second-max as we see more elements. For the multiple secretary problem, where we can pick up to r secretaries, we show constant competitiveness as long as r is large enough. For this, we give an adaptive thresholding algorithm that raises and lowers thresholds depending on the quality and the quantity of recently selected elements.
Mar. 14
  • Mert Saglam (University of Washington)
  • Abstract: We answer a 1982 conjecture of Erdős and Simonovits about the growth of number ofk-walks in a graph, which incidentally was studied even earlier by Blakley and and Dixon in 1966. We prove this conjecture in a more general setup than the earlier treatment, furthermore, through a refinement and strengthening of this inequality, we resolve two related open questions in complexity theory: the communication complexity of thek-Hamming distance isΩ(k log k)and that consequently any property tester fork-linearity requires Ω(k log k)queries.
Jan. 24
  • Mika Göös (IAS)
  • Abstract: *Separations:* We introduce a monotone variant of XOR-SAT and show it has exponential monotone circuit complexity. Since XOR-SAT is in NC^2, this improves qualitatively on the monotone vs. non-monotone separation of Tardos (1988). We also show that monotone span programs over R can be exponentially more powerful than over finite fields. These results can be interpreted as separating subclasses of TFNP in communication complexity. *Characterizations:* We show that the communication (resp. query) analogue of PPA (subclass of TFNP) captures span programs over F_2 (resp. Nullstellensatz degree over F_2). Previously, it was known that communication FP captures formulas (Karchmer--Wigderson, 1988) and that communication PLS captures circuits (Razborov, 1995). Joint work with Pritish Kamath, Robert Robere, and Dmitry Sokolov.

Related Seminars at NYU