Calibration of Shared Equilibria in General Sum Partially Observable Markov Games
arXiv:2006.13085
Abstract
Training multi-agent systems (MAS) to achieve realistic equilibria gives us a useful tool to understand and model real-world systems. We consider a general sum partially observable Markov game where agents of different types share a single policy network, conditioned on agent-specific information. This paper aims at i) formally understanding equilibria reached by such agents, and ii) matching emergent phenomena of such equilibria to real-world targets. Parameter sharing with decentralized execution has been introduced as an efficient way to train multiple agents using a single policy network. However, the nature of resulting equilibria reached by such agents has not been yet studied: we introduce the novel concept of Shared equilibrium as a symmetric pure Nash equilibrium of a certain Functional Form Game (FFG) and prove convergence to the latter for a certain class of games using self-play. In addition, it is important that such equilibria satisfy certain constraints so that MAS are calibrated to real world data for practical use: we solve this problem by introducing a novel dual-Reinforcement Learning based approach that fits emergent behaviors of agents in a Shared equilibrium to externally-specified targets, and apply our methods to a n-player market example. We do so by calibrating parameters governing distributions of agent types rather than individual agents, which allows both behavior differentiation among agents and coherent scaling of the shared policy network to multiple agents.
arXiv commentNeurIPS 2020, Thirty-fourth Conference on Neural Information Processing Systems
References in corpus (13)
- Trust Region Policy Optimization
- Learning to Communicate with Deep Multi-Agent Reinforcement Learning
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning
- A Survey and Critique of Multiagent Deep Reinforcement Learning
- Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization
- Stabilising Experience Replay for Deep Multi-Agent Reinforcement Learning
- QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning
- Learning to Optimize
- Policy Gradients with Variance Related Risk Criteria
- Pure Saddle Points and Symmetric Relative Payoff Games
- Open-ended Learning in Symmetric Zero-sum Games
- The AI Economist: Improving Equality and Productivity with AI-Driven Tax Policies
- Finite Sample Analysis of Two-Timescale Stochastic Approximation with Applications to Reinforcement Learning