This paper studies policy optimization algorithms for multi-agent reinforcement learning. We begin by proposing an algorithm framework for two-player zero-sum Markov Games in the full-information setting, where each iteration consists of a policy update step at each state using a certain matrix game algorithm, and a value update step with a certain learning rate. This framework unifies many existing and new policy optimization algorithms. We show that the state-wise average policy of this algorithm converges to an approximate Nash equilibrium (NE) of the game, as long as the matrix game algorithms achieve low weighted regret at each state, with respect to weights determined by the speed of the value updates. Next, we show that this framework instantiated with the Optimistic Follow-The-Regularized-Leader (OFTRL) algorithm at each state (and smooth value updates) can find an $\mathcal{\widetilde{O}}(T^{-5/6})$ approximate NE in $T$ iterations, and a similar algorithm with slightly modified value update rule achieves a faster $\mathcal{\widetilde{O}}(T^{-1})$ convergence rate. These improve over the current best $\mathcal{\widetilde{O}}(T^{-1/2})$ rate of symmetric policy optimization type algorithms. We also extend this algorithm to multi-player general-sum Markov Games and show an $\mathcal{\widetilde{O}}(T^{-3/4})$ convergence rate to Coarse Correlated Equilibria (CCE). Finally, we provide a numerical example to verify our theory and investigate the importance of smooth value updates, and find that using "eager" value updates instead (equivalent to the independent natural policy gradient algorithm) may significantly slow down the convergence, even on a simple game with $H=2$ layers.
翻译:本文研究多试剂强化学习的政策优化算法 。 我们首先为全信息环境下的双玩者 零和 Markov 运动会提出一个算法框架, 每个迭代包括每个州的政策更新步骤, 使用特定的矩阵游戏算法, 并使用一定的学习速度。 这个框架统一了许多现有的和新的政策优化算法 。 我们显示, 只要矩阵游戏算法在每一个州实现低加权递归率的遗憾, 由数值更新速度决定的重量。 接下来, 我们显示这个框架与每个州的最佳后续矩阵游戏算法同步( 和平滑值更新) 。 这个算法的平均政策将找到一个$mathalal_ blight_ ormaxloral_\\\\\\ 3_ 3 listal_ listalal_ laxlational_ lax a more legal_ legal_ laxal- cal- laxal- laxal laxal_ lax lax lax ladeal dal_ dal dal laxal laxal dal dal dal dal dal dal lax dal dal dal dal dal daldaldal dal date) exlations a s brox s malxxx s malxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx