Research map / Multi-agent systems and games / Cooperative multi-agent RL
Influence-Based Multi-Agent Exploration
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add EITI influence reward to PPO | PPO multi-agent exploration · Pass | team reward | Helped |
| Add EDTI Value of Interaction reward to PPO | PPO multi-agent exploration · Pass | team reward | Helped |
| Use EDTI rather than EITI to value interaction points | PPO multi-agent exploration · Secret-room | team reward | Helped |
| Add influence-based exploration rewards | PPO multi-agent exploration · Push-box | team reward | Helped |
| Add influence-based exploration rewards | PPO multi-agent exploration · Island | team reward | Helped |
| Add influence-based exploration rewards | PPO multi-agent exploration · Large-island | team reward | Helped |
| Use only immediate reward influence, without long-term transition influence (ablation) | EDTI-style PPO (r_influence) · Pass | team reward | Helped |
| Replace counterfactual interaction value with other agents' value functions (plusV) (ablation) | EDTI-style PPO · Pass | exploration efficiency | Hurt |
| Use a shared centralized critic with centralized curiosity (ablation) | PPO multi-agent exploration (shared_critic) · Sparse-reward tasks | team reward | Mixed |
| Use Q-Q without the explicit counterfactual formulation (ablation) | EDTI · Pass | team reward | No clear effect |
| Use EITI, which captures transition interactions but does not filter interactions by their value (ablation) | EDTI · Secret-room | team reward | Hurt |
| Use EITI rather than EDTI when interaction points align with extrinsic rewards | PPO multi-agent exploration · Tasks where all interaction points align with extrinsic rewards | team reward | Mixed |
Closest papers by meaning
- Individual Contributions as Intrinsic Exploration Scaffolds for Multi-agent Reinforcement Learning (ICML 2024 · similarity 0.90)
- Detecting Influence Structures in Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.89)
- Situation-Dependent Causal Influence-Based Cooperative Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.89)
- Imagine, Initialize, and Explore: An Effective Exploration Method in Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.88)
- Measuring Mutual Policy Divergence for Multi-Agent Sequential Exploration (NeurIPS 2024 · similarity 0.88)
- Settling Decentralized Multi-Agent Coordinated Exploration by Novelty Sharing (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.