Session

Oral Session 1: Theory

Moderator: Nicolò Cesa-Bianchi



Tue 7 Dec midnight PST — 1 a.m. PST

Abstract:

Chat is not available.

Tue 7 Dec. 0:00 - 0:15 PST

(Oral)
Separation Results between Fixed-Kernel and Feature-Learning Probability Metrics

Carles Domingo i Enrich · Youssef Mroueh

Several works in implicit and explicit generative modeling empirically observed that feature-learning discriminators outperform fixed-kernel discriminators in terms of the sample quality of the models. We provide separation results between probability metrics with fixed-kernel and feature-learning discriminators using the function classes $\mathcal{F}_2$ and $\mathcal{F}_1$ respectively, which were developed to study overparametrized two-layer neural networks. In particular, we construct pairs of distributions over hyper-spheres that can not be discriminated by fixed kernel $(\mathcal{F}_2)$ integral probability metric (IPM) and Stein discrepancy (SD) in high dimensions, but that can be discriminated by their feature learning ($\mathcal{F}_1$) counterparts. To further study the separation we provide links between the $\mathcal{F}_1$ and $\mathcal{F}_2$ IPMs with sliced Wasserstein distances. Our work suggests that fixed-kernel discriminators perform worse than their feature learning counterparts because their corresponding metrics are weaker.

Tue 7 Dec. 0:15 - 0:20 PST

(Q&A)
Q&A

Tue 7 Dec. 0:20 - 0:35 PST

(Oral)
Near-Optimal No-Regret Learning in General Games

Constantinos Daskalakis · Maxwell Fishelson · Noah Golowich

We show that Optimistic Hedge -- a common variant of multiplicative-weights-updates with recency bias -- attains ${\rm poly}(\log T)$ regret in multi-player general-sum games. In particular, when every player of the game uses Optimistic Hedge to iteratively update her action in response to the history of play so far, then after $T$ rounds of interaction, each player experiences total regret that is ${\rm poly}(\log T)$. Our bound improves, exponentially, the $O(T^{1/2})$ regret attainable by standard no-regret learners in games, the $O(T^{1/4})$ regret attainable by no-regret learners with recency bias (Syrgkanis et al., NeurIPS 2015), and the $O(T^{1/6})$ bound that was recently shown for Optimistic Hedge in the special case of two-player games (Chen & Peng, NeurIPS 2020). A direct corollary of our bound is that Optimistic Hedge converges to coarse correlated equilibrium in general games at a rate of $\tilde{O}(1/T)$.

Tue 7 Dec. 0:35 - 0:40 PST

(Q&A)
Q&A

Tue 7 Dec. 0:40 - 0:55 PST

(Oral)
Lower Bounds on Metropolized Sampling Methods for Well-Conditioned Distributions

Yin Tat Lee · Ruoqi Shen · Kevin Tian

We give lower bounds on the performance of two of the most popular sampling methods in practice, the Metropolis-adjusted Langevin algorithm (MALA) and multi-step Hamiltonian Monte Carlo (HMC) with a leapfrog integrator, when applied to well-conditioned distributions. Our main result is a nearly-tight lower bound of $\widetilde{\Omega}(\kappa d)$ on the mixing time of MALA from an exponentially warm start, matching a line of algorithmic results \cite{DwivediCW018, ChenDWY19, LeeST20a} up to logarithmic factors and answering an open question of \cite{ChewiLACGR20}. We also show that a polynomial dependence on dimension is necessary for the relaxation time of HMC under any number of leapfrog steps, and bound the gains achievable by changing the step count. Our HMC analysis draws upon a novel connection between leapfrog integration and Chebyshev polynomials, which may be of independent interest.

Tue 7 Dec. 0:55 - 1:00 PST

(Q&A)
Q&A