Theoretical and Practical Advances on Smoothing for Extensive-Form Games

TL;DR

This paper introduces a novel weighting scheme for the dilated entropy function, significantly improving the convergence rates of first-order methods in solving large-scale extensive-form games and surpassing counterfactual regret minimization in practice.

Contribution

It develops a new distance-generating function with no dependence on the branching factor, enhancing first-order methods' efficiency in extensive-form games.

Findings

New weighting scheme improves convergence rate by a factor of Ω(b^dd).

First-order methods can outperform CFR+ with the new approach.

Practical tuning makes the excessive gap technique faster than CFR+.

Abstract

Sparse iterative methods, in particular first-order methods, are known to be among the most effective in solving large-scale two-player zero-sum extensive-form games. The convergence rates of these methods depend heavily on the properties of the distance-generating function that they are based on. We investigate the acceleration of first-order methods for solving extensive-form games through better design of the dilated entropy function---a class of distance-generating functions related to the domains associated with the extensive-form games. By introducing a new weighting scheme for the dilated entropy function, we develop the first distance-generating function for the strategy spaces of sequential games that has no dependence on the branching factor of the player. This result improves the convergence rate of several first-order methods by a factor of $\Omega(b^dd)$, where $b$ is the branching factor of the player, and $d$ is the depth of the game tree. Thus far, counterfactual regret minimization methods have been faster in practice, and more popular, than first-order methods despite their theoretically inferior convergence rates. Using our new weighting scheme and practical tuning we show that, for the first time, the excessive gap technique can be made faster than the fastest counterfactual regret minimization algorithm, CFR+, in practice.