Research map / Multi-agent systems and games / Cooperative multi-agent RL
Partial Action Replacement: Tackling Distribution Shift in Offline MARL
Yue Jin, Giovanni Montana
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| use SPaCQL partial-action-replacement backup instead of CFCQL joint-action update | SPaCQL · MPE World Random | normalized score: 68 -> 94.3 | Improved |
| use SPaCQL instead of CFCQL | SPaCQL · MPE Cooperative Navigation Medium-Replay | normalized score: 52.2 -> 71.9 | Improved |
| use SPaCQL instead of CFCQL | SPaCQL · MaMujoco Half-Cheetah Random | normalized score: 39.7 -> 43.8 | Improved |
| use SPaCQL instead of CFCQL on expert data | SPaCQL · MPE expert datasets | normalized score: CN 112 -> 111.9; PP 118.2 -> 111.2; World 119.7 -> 112.3 | Hurt |
| use SPaCQL instead of CFCQL on medium data | SPaCQL · Medium datasets | normalized score: CN 65.0 -> 78.6; PP 68.5 -> 61.9; Half-C 80.5 -> 70.3 | Mixed |
| use ICQL-QS (single-agent replacement) instead of CFCQL on random data (ablation) | ICQL-QS · MPE Cooperative Navigation Random | normalized score: 62.2 -> 77.7 | Improved |
| use ICQL-QS instead of CFCQL on expert data (ablation) | ICQL-QS · MPE Cooperative Navigation Expert | normalized score: 112 -> 97.2 | Hurt |
| partial action replacement lowers Q-ensemble uncertainty vs joint-action update (ablation) | ICQL-QS · Cooperative Navigation | Q-ensemble std | Improved |
Closest papers by meaning
- Offline Multi-Agent Reinforcement Learning via In-Sample Sequential Policy Optimization (AAAI 2025 · similarity 0.90)
- ComaDICE: Offline Cooperative Multi-Agent Reinforcement Learning with Stationary Distribution Shift Regularization (ICLR 2025 · similarity 0.90)
- Dispelling the Mirage of Progress in Offline MARL through Standardised Baselines and Evaluation (NeurIPS 2024 · similarity 0.90)
- Revisiting Cooperative Off-Policy Multi-Agent Reinforcement Learning (ICML 2025 · similarity 0.89)
- Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.88)
- Adaptive Neighborhood-Constrained Q Learning for Offline Reinforcement Learning (NeurIPS 2025 · 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.
