Learning Policy Representations in Multiagent Systems
arXiv:1806.06464
Abstract
Modeling agent behavior is central to understanding the emergence of complex phenomena in multiagent systems. Prior work in agent modeling has largely been task-specific and driven by hand-engineering domain-specific prior knowledge. We propose a general learning framework for modeling agent behavior in any multiagent system using only a handful of interaction data. Our framework casts agent modeling as a representation learning problem. Consequently, we construct a novel objective inspired by imitation learning and agent identification and design an algorithm for unsupervised learning of representations of agent policies. We demonstrate empirically the utility of the proposed framework in (i) a challenging high-dimensional competitive environment for continuous control and (ii) a cooperative environment for communication, on supervised predictive tasks, unsupervised clustering, and policy optimization using deep reinforcement learning.
ICML 2018
References in corpus (1)
Cited by in corpus (24)
- A Survey and Critique of Multiagent Deep Reinforcement Learning
- The Hanabi Challenge: A New Frontier for AI Research
- Value Functions Factorization with Latent State Information Sharing in Decentralized Multi-Agent Policy Gradients
- ToM2C: Target-oriented Multi-agent Communication and Cooperation with Theory of Mind
- Ease-of-Teaching and Language Structure from Emergent Communication
- Learning Latent Representations to Influence Multi-Agent Interaction
- Towards More Human-like AI Communication: A Review of Emergent Communication Research
- A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning
- Towards Deployment of Robust AI Agents for Human-Machine Partnerships
- Model-Based Opponent Modeling
- Relational Graph Learning for Crowd Navigation
- Agent Modelling under Partial Observability for Deep Reinforcement Learning
- Learning from Imperfect Annotations
- Towards Open Ad Hoc Teamwork Using Graph-based Policy Learning
- Multi-Agent Interactions Modeling with Correlated Policies
- Learning Complex Multi-Agent Policies in Presence of an Adversary
- Some performance considerations when using multi-armed bandit algorithms in the presence of missing data
- General Characterization of Agents by States they Visit
- Learning Altruistic Behaviours in Reinforcement Learning without External Rewards
- Dynamic Relational Inference in Multi-Agent Trajectories
- Classification Under Human Assistance
- Multi-agent Policy Optimization with Approximatively Synchronous Advantage Estimation
- Agent Probing Interaction Policies
- Regression Under Human Assistance