ICLR 2023 accepted papers on Multi-agent systems and games
37 of the 1,575 papers accepted at ICLR 2023 (2.3%) are on Multi-agent systems and games, compared with 3.0% at ICLR 2022. They fall into 3 topics.
See ICLR 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Human-AI and goal-conditioned · 17 papers
- Leveraging Future Relationship Reasoning for Vehicle Trajectory Prediction (18 citations)
- Emergence of Maps in the Memories of Blind Navigation Agents (17 citations)
- Transformers are Sample-Efficient World Models (12 citations)
- Human-level Atari 200x faster (6 citations)
- Computational Language Acquisition with Theory of Mind (5 citations)
- Learning Zero-Shot Cooperation with Humans, Assuming Humans Are Biased (4 citations)
- Towards Effective and Interpretable Human-Agent Collaboration in MOBA Games: A Communication Perspective (2 citations)
- Neural Agents Struggle to Take Turns in Bidirectional Emergent Communication (1 citation)
- Evaluating Long-Term Memory in 3D Mazes
- Moving Forward by Moving Backward: Embedding Action Impact over Action Semantics
- Relative Behavioral Attributes: Filling the Gap between Symbolic Goal Specification and Reward Learning from Human Preferences
- Adversarial Diversity in Hanabi
and 5 more in the interactive map.
Cooperative multi-agent RL · 12 papers
- More Centralized Training, Still Decentralized Execution: Multi-Agent Conditional Policy Factorization (5 citations)
- Order Matters: Agent-by-agent Policy Optimization (3 citations)
- Solving Continuous Control via Q-learning (3 citations)
- Stateful Active Facilitator: Coordination and Environmental Heterogeneity in Cooperative Multi-Agent Reinforcement Learning (1 citation)
- Cheap Talk Discovery and Utilization in Multi-Agent Reinforcement Learning
- Boosting Multiagent Reinforcement Learning via Permutation Invariant and Permutation Equivariant Networks
- Certifiably Robust Policy Learning against Adversarial Multi-Agent Communication
- Discovering Generalizable Multi-agent Coordination Skills from Multi-task Offline Data
- Generating Diverse Cooperative Agents by Learning Incompatible Policies
- GoBigger: A Scalable Platform for Cooperative-Competitive Multi-Agent Interactive Simulation
- MACTA: A Multi-agent Reinforcement Learning Approach for Cache Timing Attacks and Detection
- RPM: Generalizable Multi-Agent Policies for Multi-Agent Reinforcement Learning
Games and equilibria · 8 papers
- Mastering the Game of No-Press Diplomacy via Human-Regularized Reinforcement Learning and Planning (11 citations)
- MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning (3 citations)
- Scaling Laws for a Multi-Agent Reinforcement Learning Model (3 citations)
- Efficiently Computing Nash Equilibria in Adversarial Team Markov Games (2 citations)
- Faster Last-iterate Convergence of Policy Optimization in Zero-Sum Markov Games (2 citations)
- Asynchronous Gradient Play in Zero-Sum Multi-agent Games
- Population-size-Aware Policy Optimization for Mean-Field Games
- Truthful Self-Play