Research map / Multi-agent systems and games / Cooperative multi-agent RL
Weighted QMIX: Expanding Monotonic Value Function Factorisation for Deep 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 |
|---|---|---|---|
| add weighting (CW or OW) plus unrestricted Q_central to QMIX | QMIX · Predator Prey (8 agents) | median test return | Helped |
| add unrestricted Q_central only (uniform weighting) (ablation) | QMIX · Predator Prey (8 agents) | median test return | No clear effect |
| add CW/OW weighting only (no Q_central) (ablation) | QMIX · Predator Prey (8 agents) | median test return | No clear effect |
| add weighting plus Q_central under increased exploration | QMIX · SMAC 3s5z and 5m_vs_6m (eps annealed over 1M steps) | median test win % | Helped |
| add weighting only to QMIX under increased exploration (ablation) | QMIX · SMAC 5m_vs_6m | median test win % | Hurt |
| add weighting only to QMIX under increased exploration (ablation) | QMIX · SMAC 3s5z | median test win % | No clear effect |
| use Weighted QMIX instead of QMIX | QMIX · SMAC bane_vs_bane | median test win % | Helped |
| use Weighted QMIX instead of QMIX | QMIX · SMAC 6h_vs_8z (1M-step epsilon) | median test win % | Helped |
| use Weighted QMIX instead of QMIX | QMIX · SMAC corridor | median test win % | Hurt |
| use Weighted QMIX instead of QMIX | QMIX · SMAC 3s5z_vs_3s6z | median test win % | No clear effect |
| modify Q_central architecture (ablation) | Weighted QMIX · SMAC corridor | median test win % | Helped |
| set weighting alpha too low (ablation) | CW-QMIX and OW-QMIX · SMAC (increased exploration) | median test win % | Hurt |
Closest papers by meaning
- QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning (ICML 2018 · similarity 0.95)
- Regularized Softmax Deep Multi-Agent Q-Learning (NeurIPS 2021 · similarity 0.91)
- Potentially Optimal Joint Actions Recognition for Cooperative Multi-Agent Reinforcement Learning (ICLR 2026 · similarity 0.91)
- QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement Learning (ICML 2019 · similarity 0.91)
- ConcaveQ: Non-monotonic Value Function Factorization via Concave Representations in Deep Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.89)
- Learning Joint Behaviors with Large Variations (AAAI 2025 · similarity 0.87)
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.