Research map / Multi-agent systems and games / Cooperative multi-agent RL
Offline Multi-Agent Reinforcement Learning with Implicit Global-to-Local Value Regularization
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 implicit global-to-local value regularization (OMIGA) instead of local-level regularization (ICQ) | offline MARL on multi-agent MuJoCo · Hopper medium | average return: 501.79 -> 1189.26 | Helped |
| OMIGA versus best baseline on SMAC poor and medium-expert datasets | OMIGA · 5m_vs_6m poor / HalfCheetah medium-expert | score: 7.61 -> 7.52 (5m_vs_6m poor); 3543.70 -> 2948.46 (HalfCheetah medium-expert) | Hurt |
| use OMIGA on mixed-quality (heterogeneous behavior) datasets versus best baseline | OMIGA · SMAC corridor medium-poor | average score: 8.27 -> 9.70 | Helped |
| add global information (global weight w_i(o)) to local policy learning (ablation) | OMIGA · HalfCheetah and 6h_vs_8z datasets | performance and stability | Helped |
| vary regularization alpha (too high or too low) (ablation) | OMIGA · HalfCheetah | performance | Hurt |
| increase alpha (stronger regularization) (ablation) | OMIGA · SMAC 6h_vs_8z | performance | Helped |
Closest papers by meaning
- ComaDICE: Offline Cooperative Multi-Agent Reinforcement Learning with Stationary Distribution Shift Regularization (ICLR 2025 · similarity 0.93)
- Offline Multi-Agent Reinforcement Learning via In-Sample Sequential Policy Optimization (AAAI 2025 · similarity 0.91)
- Dispelling the Mirage of Progress in Offline MARL through Standardised Baselines and Evaluation (NeurIPS 2024 · similarity 0.90)
- A Clean Slate for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.90)
- Exploiting Structure in Offline Multi-Agent RL: The Benefits of Low Interaction Rank (ICLR 2025 · similarity 0.89)
- Scaling Offline Model-Based RL via Jointly-Optimized World-Action Model Pretraining (ICLR 2025 · similarity 0.89)
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.