Research map / Multi-agent systems and games / Cooperative multi-agent RL
DeCOM: Decomposed Policy for Constrained Cooperative Multi-Agent Reinforcement Learning
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 |
|---|---|---|---|
| Use decomposed policies with shared neighboring agents' base actions (ablation) | DeCOM-A · CLFM | Revenue: 19004.70 $\pm$ 76.88 -> 19286.68 $\pm$ 78.42 | Helped |
| Use decomposed policies with shared neighboring agents' base actions (ablation) | DeCOM-A · CLFM | Demand-Supply Gap: 97.33 $\pm$ 1.28 -> 89.11 $\pm$ 3.53 | Helped |
| Use neighboring agents' base actions rather than no communication (ablation) | DeCOM-A · CLFM | Revenue: 18668.26 $\pm$ 362.78 -> 19286.68 $\pm$ 78.42 | Helped |
| Use neighboring agents' base actions rather than no communication (ablation) | DeCOM-A · CLFM | Demand-Supply Gap: 105.57 $\pm$ 5.06 -> 89.11 $\pm$ 3.53 | Helped |
| Use neighboring agents' base actions rather than no communication (ablation) | DeCOM-A · CTC-fair | Unfairness: 9.22 $\pm$ 0.78 -> 8.11 $\pm$ 0.08 | Helped |
| Use neighboring agents' base actions rather than no communication (ablation) | DeCOM-A · CTC-fair | Reward: 4.36 $\pm$ 0.65 -> 3.76 $\pm$ 0.84 | Hurt |
| Use neighboring agents' base actions rather than no communication (ablation) | DeCOM-A · CDSN | Reward: 51.93 $\pm$ 24.05 -> 58.41 $\pm$ 11.21 | Helped |
| Use neighboring agents' base actions rather than no communication (ablation) | DeCOM-A · CDSN | Operational Cost: 32.88 $\pm$ 4.25 -> 58.90 $\pm$ 26.23 | Hurt |
| Add neighboring base actions to the perturbation-policy inputs (ablation) | DeCOM-A · CTC-safe | Unsafety 1: 0.43 $\pm$ 0.28 -> 0.43 $\pm$ 0.11 | No clear effect |
| Use only local observation as the perturbation-policy input (ablation) | DeCOM-I · CTC-safe | Unsafety 1: 0.43 $\pm$ 0.11 -> 0.38 $\pm$ 0.09 | Helped |
| Use only local observation as the perturbation-policy input (ablation) | DeCOM-I · CTC-fair | Unfairness: 8.11 $\pm$ 0.08 -> 8.05 $\pm$ 0.20 | No clear effect |
| Use only local observation as the perturbation-policy input (ablation) | DeCOM-I · CLFM | Revenue: 19286.68 $\pm$ 78.42 -> 19004.70 $\pm$ 76.88 | Hurt |
Closest papers by meaning
- Cooperative Policy Agreement: Learning Diverse Policy for Offline MARL (AAAI 2025 · similarity 0.89)
- Decomposing Temporal Equilibrium Strategy for Coordinated Distributed Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.89)
- Learning Joint Behaviors with Large Variations (AAAI 2025 · similarity 0.89)
- ComaDICE: Offline Cooperative Multi-Agent Reinforcement Learning with Stationary Distribution Shift Regularization (ICLR 2025 · similarity 0.88)
- Offline Multi-Agent Reinforcement Learning via In-Sample Sequential Policy Optimization (AAAI 2025 · similarity 0.88)
- Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems (NeurIPS 2026 · 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.