Causal Multi-Agent Reinforcement Learning: Review and Open Problems

Abstract

This paper serves to introduce the reader to the field of multi-agent reinforcement learning (MARL) and its intersection with methods from the study of causality. We highlight key challenges in MARL and discuss these in the context of how causal methods may assist in tackling them. We promote moving toward a 'causality first' perspective on MARL. Specifically, we argue that causality can offer improved safety, interpretability, and robustness, while also providing strong theoretical guarantees for emergent behaviour. We discuss potential solutions for common challenges, and use this context to motivate future research directions.

Accepted at Cooperative AI Workshop, NeurIPS 2021

References in corpus (12)

• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• Dealing with Non-Stationarity in Multi-Agent Deep Reinforcement Learning
• Causality for Machine Learning
• Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data
• Causal Imitation Learning with Unobserved Confounders
• Causal Reinforcement Learning using Observational and Interventional Data
• Scalable Multi-Agent Reinforcement Learning for Networked Systems with Average Reward
• On Inductive Biases in Deep Reinforcement Learning
• Meta-Reinforcement Learning Robust to Distributional Shift via Model Identification and Experience Relabeling
• Believe What You See: Implicit Constraint Approach for Offline Multi-Agent Reinforcement Learning
• Causal Markov Decision Processes: Learning Good Interventions Efficiently
• A Critical View of the Structural Causal Model