Research map / Multi-agent systems and games / Cooperative multi-agent RL
Context-Aware Sparse Deep Coordination Graphs
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 |
|---|---|---|---|
| Select context-aware sparse edges using payoff variance instead of using a complete coordination graph (ablation) | CASEC · SMAC: 5m_vs_6m and MMM2 | sample efficiency | Helped |
| Use action representations instead of conventional Q networks (ablation) | CASEC · SMAC: 5m_vs_6m | training stability | Helped |
| Use action representations instead of conventional Q networks (ablation) | CASEC · SMAC | TD error | Helped |
| Use action representations instead of low-rank approximation (ablation) | CASEC payoff learning · SMAC | TD error | Helped |
| Remove the sparseness loss (ablation) | CASEC · Sensor | reward | Hurt |
| Use a complete graph rather than context-aware sparse topologies (ablation) | CASEC · Sensor | number of scanned targets | Hurt |
| Use a complete graph rather than context-aware sparse topologies (ablation) | CASEC · SMAC | sample efficiency | Hurt |
| Use payoff variance to construct sparse graphs rather than observation-based attention (ablation) | CASEC · MACO | performance | Helped |
| Use payoff variance to construct sparse graphs rather than observation-based attention (ablation) | CASEC · Disperse | performance | Hurt |
| Use a sparse adaptive graph rather than DCG's complete static graph | CASEC · SMAC: 5m_vs_6m and MMM2 | performance | Helped |
| Use a sparse adaptive graph rather than DCG's complete static graph | CASEC · Sensor | number of scanned targets | Helped |
| Use payoff-variance graph construction rather than fully decomposed value learning | CASEC · MACO | performance | Helped |
Closest papers by meaning
- Graph Diffusion for Robust Multi-Agent Coordination (ICML 2025 · similarity 0.90)
- Learning Mean Field Games on Sparse Graphs: A Hybrid Graphex Approach (ICLR 2024 · similarity 0.87)
- Bridging Training and Execution via Dynamic Directed Graph-Based Communication in Cooperative Multi-Agent Systems (AAAI 2025 · similarity 0.86)
- Learning Mean Field Control on Sparse Graphs (ICML 2025 · similarity 0.86)
- Bayesian Ego-graph inference for Networked Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- Towards Efficient Collaboration via Graph Modeling in Reinforcement Learning (AAAI 2025 · similarity 0.86)
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.