- You are here:
- GT Home
- Home
- News & Events

Series: ACO Student Seminar

We resolve an open question from (Christiano, 2014b) posed in COLT'14 regarding the optimal dependency of the regret achievable for online local learning on the size of the label set. In this framework the algorithm is shown a pair of items at each step, chosen from a set of n items. The learner then predicts a label for each item, from a label set of size L and receives a real valued payoff. This is a natural framework which captures many interesting scenarios such as collaborative filtering, online gambling, and online max cut among others. (Christiano, 2014a) designed an efficient online learning algorithm for this problem achieving a regret of O((nL^3T)^(1/2)), where T is the number of rounds. Information theoretically, one can achieve a regret of O((n log LT)^(1/2)). One of the main open questions left in this framework concerns closing the above gap.
In this work, we provide a complete answer to the question above via two main results. We show, via a tighter analysis, that the semi-definite programming based algorithm of (Christiano, 2014a), in fact achieves a regret of O((nLT)^(1/2)). Second, we show a matching computational lower bound. Namely, we show that a polynomial time algorithm for online local learning with lower regret would imply a polynomial time algorithm for the planted clique problem which is widely believed to be hard. We prove a similar hardness result under a related conjecture concerning planted dense subgraphs that we put forth. Unlike planted clique, the planted dense subgraph problem does not have any known quasi-polynomial time algorithms. Computational lower bounds for online learning are relatively rare, and we hope that the ideas developed in this work will lead to lower bounds for other online learning scenarios as well.
Joint work with Pranjal Awasthi, Moses Charikar, and Andrej Risteski at Princeton.

Series: ACO Student Seminar

Yannakakis showed that the matching problem does not have a small symmetric linear program. Rothvoß recently proved that any (not necessarily symmetric) linear program also has exponential size. It is natural to ask whether the matching problem can be expressed compactly in a framework such as semidefinite programming (SDP) that is more powerful than linear programming but still allows efficient optimization. We answer this question negatively for symmetric SDPs: any symmetric SDP for the matching problem has exponential size. We also show that an O(k)-round Lasserre SDP relaxation for the metric traveling salesperson problem (TSP) yields at least as good an approximation as any symmetric SDP relaxation of size n^k. The key technical ingredient underlying both these results is an upper bound on the degree needed to derive polynomial identities that hold over the space of matchings or traveling salesperson tours.
This is joint work with Jonah Brown-Cohen, Prasad Raghavendra and Benjamin Weitz from Berkeley, and Gabor Braun, Sebastian Pokutta, Aurko Roy and Daniel Zink at Georgia Tech.

Series: ACO Student Seminar

Many statistical physics models are defined on an infinite lattice by taking appropriate limits of the model on finite lattice regions. A key consideration is which boundary to use when taking these limits, since the boundary can have significant influence on properties of the limit. Fixed boundary conditions assume that the boundary cells are given a fixed assignment, and free boundary conditions allow these cells to vary, taking the union of all possible fixed boundaries. It is known that these two boundary conditions can cause significant differences in physical properties, such as whether there is a phase transition, as well as computational properties, including whether local Markov chain algorithms used to sample and approximately count are efficient.
We consider configurations with free or partially free boundary conditions and show that by randomly extending the boundary by a few layers, choosing among only a constant number of allowable extensions, we can generalize the arguments used in the fixed boundary setting to infer bounds on the mixing time for free boundaries. We demonstrate this principled approach using randomized extensions for 3-colorings of regions of Z2 and lozenge tilings of regions of the triangle lattice, building on arguments for the fixed boundary cases due to Luby et.al. Our approach yields an efficient algorithm for sampling free boundary 3-colorings of regions with one reflex corner, the first result to efficiently sample free boundary 3-colorings of any nonconvex region. We also consider self-reducibility of free boundary 3-colorings of rectangles, and show our algorithm can be used to approximately count the number of free-boundary 3-colorings of a rectangle.

Series: ACO Student Seminar

A graph is a minor of another graph if the former can be obtained from a subgraph of the latter by contracting edges. We prove that for every graph H, if H is not a minor of a graph G, then V(G) can be 3-colored such that the subgraph induced by each color class has no component with size greater than a function of H and the maximum degree of G. This answers a question raised by Esperet and Joret, generalizes their result for 3-coloring V(G) for graphs G embeddable in a fixed surface, and improves a result of Alon, Ding, Oporowski and Vertigan for 4-coloringing V(G) for H-minor free graphs G. As a corollary, we prove that for every positive integer t, if G is a graph with no K_{t+1} minor, then V(G) can be 3t-colored such that the subgraph induced by each color class has no component with size larger than a function of t. This improves a result of Wood for coloring V(G) by 3.5t+2 colors. This work is joint with Sang-il Oum.

Series: ACO Student Seminar

How much of space can be filled with pairwise non-overlapping copies of a given solid? This is one of the oldest problems in mathematics, intriguing since the times of Aristotle, and remaining remarkably elusive until present times. For example, the three-dimensional sphere packing problem (posed by Kepler in 1611) was only solved in 1998 by Ferguson and Hales.
In this talk, I will provide some historical and modern applications of geometric packing problems, and I will introduce a methodology to derive upper bounds on the maximal density of such packings. These upper bounds are obtained by an infinite dimensional linear program, which is not computationally tractable. However, this problem can be approximated by using tools from sums of squares relaxations and symmetry reduction (harmonic analysis and representation theory), leading to rigorous
computational upper bounds on the density.
Time permitting, I will present ongoing work with Maria Dostert, Fernando de Oliveira Filho and Frank Vallentin on the density of translative packings of superspheres (i.e., ell_p balls).
This is an introductory talk: no previous knowledge of sums of squares relaxations or symmetry reduction is assumed.

Series: ACO Student Seminar

Joint ARC colloquium/ACO student seminar

Pursuit games---motivated historically by military tactics---are
a natural for graphical settings, and take many forms. We will
present some recent results involving (among other things) drunks,
Kakeya sets and a "ketchup graph.'' Lastly, we describe what we
think is the most important open problem in the field.

Series: ACO Student Seminar

Given a family of ideals which are symmetric under some group action on the variables, a natural question to ask is whether the generating set stabilizes up to symmetry as the number of variables tends to infinity. We answer this in the affirmative for a broad class of toric ideals, settling several open questions coming from algebraic statistics. Our approach involves factoring an equivariant monomial map into a part for which we have an explicit degree bound of the kernel, and a part for which we canprove that the source, a so-called matching monoid, is equivariantly Noetherian. The proof is mostly combinatorial, making use of the theory of well-partial orders and its relationship to Noetherianity of monoid rings. Joint work with Jan Draisma, Rob Eggermont, and Anton Leykin.

Series: ACO Student Seminar

Database privacy has garnered a recent surge in interest from the theoretical science community following the seminal work of Dwork 2006, which proposed the strong notion of differential privacy. In this setting, each row of a database corresponds to the data owned by some (distinct) individual. An analyst submits a database query to a differentially private mechanism, which replies with a noisy answer guaranteeing privacy for the data owners and accuracy for the analyst. The mechanism's privacy parameter \epsilon is correlated negatively with privacy and positively with accuracy.This work builds a framework for creating and analyzing a market that 1) solves for some socially efficient value of \epsilon using the privacy and accuracy preferences of a heterogeneous body of data owners and a single analyst, 2) computes a noisy statistic on the database, and 3) collects and distributes payments for privacy that elicit truthful reporting of data owners' preferences. We present a market for database privacy in this new framework expanding on the public goods market of Groves and Ledyard, 1977.

Series: ACO Student Seminar

We present randomized algorithms for sampling the standard Gaussian distribution restricted to a convex set and for estimating the Gaussian measure of a convex set, in the general membership oracle model. The complexity of the integration algorithm is O*(n^3) while the complexity of the sampling algorithm is O*(n^3) for the first sample and O*(n^2) for every subsequent sample. These bounds improve on the corresponding state-of-the-art by a factor of n. Our improvement comes from several aspects: better isoperimetry, smoother annealing, avoiding transformation to isotropic position and the use of the ``speedy walk" in the analysis.

Series: ACO Student Seminar

Since the 50s and Nash general proof of equilibrium existence in games it
is well understood
that even simple games may have many, even uncountably infinite, equilibria
with different properties.
In such cases a natural question arises, which equilibrium is the right one?
In this work, we perform average case analysis of evolutionary dynamics in
such cases of games.
Intuitively, we assign to each equilibrium a probability mass that is
proportional to the size of its
region of attraction. We develop new techniques to compute these
likelihoods for classic games
such as the Stag Hunt game (and generalizations) as well as balls-bins
games. Our proofs combine
techniques from information theory (relative entropy), dynamical systems
(center manifold theorem),
and algorithmic game theory.
Joint work with Georgios Piliouras