Shapley Counterfactual Credits for Multi-Agent Reinforcement Learning

Abstract

Centralized Training with Decentralized Execution (CTDE) has been a popular paradigm in cooperative Multi-Agent Reinforcement Learning (MARL) settings and is widely used in many real applications. One of the major challenges in the training process is credit assignment, which aims to deduce the contributions of each agent according to the global rewards. Existing credit assignment methods focus on either decomposing the joint value function into individual value functions or measuring the impact of local observations and actions on the global value function. These approaches lack a thorough consideration of the complicated interactions among multiple agents, leading to an unsuitable assignment of credit and subsequently mediocre results on MARL. We propose Shapley Counterfactual Credit Assignment, a novel method for explicit credit assignment which accounts for the coalition of agents. Specifically, Shapley Value and its desired properties are leveraged in deep MARL to credit any combinations of agents, which grants us the capability to estimate the individual credit for each agent. Despite this capability, the main technical difficulty lies in the computational complexity of Shapley Value who grows factorially as the number of agents. We instead utilize an approximation method via Monte Carlo sampling, which reduces the sample complexity while maintaining its effectiveness. We evaluate our method on StarCraft II benchmarks across different scenarios. Our method outperforms existing cooperative MARL algorithms significantly and achieves the state-of-the-art, with especially large margins on tasks with more severe difficulties.

References in corpus (13)

• Continuous control with deep reinforcement learning
• Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
• Counterfactual Multi-Agent Policy Gradients
• Mean Field Multi-Agent Reinforcement Learning
• Actor-Attention-Critic for Multi-Agent Reinforcement Learning
• The Complexity of Decentralized Control of Markov Decision Processes
• Data Shapley: Equitable Valuation of Data for Machine Learning
• Problems with Shapley-value-based explanations as feature importance measures
• Qatten: A General Framework for Cooperative Multiagent Reinforcement Learning
• Shapley Q-value: A Local Reward Approach to Solve Global Reward Games
• Neuron Shapley: Discovering the Responsible Neurons
• Shapley Flow: A Graph-based Approach to Interpreting Model Predictions
• Q-value Path Decomposition for Deep Multiagent Reinforcement Learning

Cited by in corpus (4)

• Instance-wise or Class-wise? A Tale of Neighbor Shapley for Concept-based Explanation
• Replication-Robust Payoff-Allocation for Machine Learning Data Markets
• A Comprehensive Study of Shapley Value in Data Analytics
• Quantifying Feature Importance of Games and Strategies via Shapley Values