When solving two-player zero-sum games, multi-agent reinforcement learning (MARL) algorithms often create populations of agents where, at each iteration, a new agent is discovered as the best response to a mixture over the opponent population. Within such a process, the update rules of "who to compete with" (i.e., the opponent mixture) and "how to beat them" (i.e., finding best responses) are underpinned by manually developed game theoretical principles such as fictitious play and Double Oracle. In this paper, we introduce a novel framework -- Neural Auto-Curricula (NAC) -- that leverages meta-gradient descent to automate the discovery of the learning update rule without explicit human design. Specifically, we parameterise the opponent selection module by neural networks and the best-response module by optimisation subroutines, and update their parameters solely via interaction with the game engine, where both players aim to minimise their exploitability. Surprisingly, even without human design, the discovered MARL algorithms achieve competitive or even better performance with the state-of-the-art population-based game solvers (e.g., PSRO) on Games of Skill, differentiable Lotto, non-transitive Mixture Games, Iterated Matching Pennies, and Kuhn Poker. Additionally, we show that NAC is able to generalise from small games to large games, for example training on Kuhn Poker and outperforming PSRO on Leduc Poker. Our work inspires a promising future direction to discover general MARL algorithms solely from data.
翻译:当解决双玩者零和游戏时,多试剂强化学习(MARL)算法(MARL)通常会创造代理商群,在每次迭代时,都会发现一个新的代理商,这是对对手群中混合的最好反应。在这一过程中,我们通过“谁竞争”(即对手混合)和“如何击败他们”(即找到最佳反应)更新规则,这得到人工开发的游戏理论原则的支持,例如假游戏和双甲骨文等。在本文中,我们引入了一个新的框架 -- -- 神经自动库(NAC) -- -- 利用基因梯级的下降将学习更新规则的发现自动化,而没有明确的人类设计。具体来说,我们通过神经网络和最佳反应模块对“谁竞争”(即对手混合)和“如何击败他们”(即找到最佳反应)规则进行调整,并且仅仅通过与游戏引擎互动来更新其参数,让两个玩家都力求最大限度地减少其利用性。令人惊讶的是,即使没有人性设计,我们所发现的MARL的预感知的预感(ML)的预感),在S-Mod-mocial-Model-model-model-model-model-model-model-model-model-model-model-model-model-model-mocial-mocial-model-model-model-model-mocal-mocal-mocal-model-model-model-model-model-model-modal-model-model-modia-modia-modia-model-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-modia-mod-mod-mod-mod-mod-mod-modia-modia-modia-modia-modia-modia-modia-modia-mod