Research map / Multi-agent systems and games / Cooperative multi-agent RL
Q-value Path Decomposition for Deep Multiagent 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 QPD instead of QMIX | Cooperative MARL on SMAC · SMAC 3s5z | median win percentage: 16 -> 85 | Helped |
| Use QPD instead of QMIX | Cooperative MARL on SMAC · SMAC 3s5z_vs_3s6z | median win percentage: 0 -> 8 | Helped |
| Use QPD instead of QMIX | Cooperative MARL on SMAC · SMAC 1c3s5z | median win percentage: 89 -> 92 | Helped |
| Use QPD instead of the strongest median baseline | Cooperative MARL on SMAC · 3m, 8m, and 2s3z | median win percentage: 100 -> 95; 100 -> 94; 100 -> 95 | Hurt |
| Use a modular multi-channel critic instead of a naive fully connected critic (ablation) | QPD · StarCraft II micromanagement tasks | win percentage | Helped |
| Use concatenation to integrate hidden features from channels (ablation) | QPD multi-channel critic · StarCraft II micromanagement tasks | win percentage | Helped |
| Increase the decomposition step when the step count is low (ablation) | QPD · StarCraft II micromanagement tasks | win percentage | Helped |
| Use 5 decomposition steps rather than 10 or 25 (ablation) | QPD · StarCraft II micromanagement tasks | win percentage | No clear effect |
Closest papers by meaning
- Revisiting Some Common Practices in Cooperative Multi-Agent Reinforcement Learning (ICML 2022 · similarity 0.91)
- Towards Understanding Cooperative Multi-Agent Q-Learning with Value Factorization (NeurIPS 2021 · similarity 0.91)
- Multiagent Q-learning with Sub-Team Coordination (NeurIPS 2022 · similarity 0.91)
- Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Tasks (NeurIPS 2021 · similarity 0.90)
- Gradient-Protected Value Decomposition for Cooperative Multi-Agent Reinforcement Learning (AAAI 2026 · similarity 0.90)
- Locality Matters: A Scalable Value Decomposition Approach for Cooperative Multi-Agent Reinforcement Learning (AAAI 2022 · similarity 0.90)
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.