Research map / Multi-agent systems and games / Cooperative multi-agent RL
Integrated Cooperation and Competition in Multi-Agent Decision-Making
Kyle Wray, Akshat Kumar, Shlomo Zilberstein
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| increase controller nodes per agent from 2 to 4 | CCP approximate algorithm · Battle Meeting | total computation time (sec.): 447.2 -> 2827.3 | Hurt |
| increase controller nodes per agent from 4 to 6 | CCP approximate algorithm · Battle Meeting | total computation time (sec.): 2827.3 -> 10775.3 | Hurt |
| increase controller nodes per agent from 2 to 4 | CCP approximate algorithm · Prisoner Meeting | total computation time (sec.): 574.1 -> 2577.0 | Hurt |
| increase controller nodes per agent from 4 to 6 | CCP approximate algorithm · Prisoner Meeting | total computation time (sec.): 2577.0 -> 11702.8 | Hurt |
| increase controller nodes from 2 to 6 | CCP approximate algorithm · Battle Meeting | best response iterations: 10.5 -> 16.5 | Hurt |
| increase controller nodes from 2 to 6 | CCP approximate algorithm · Prisoner Meeting | best response iterations: 13.3 -> 38.3 | Hurt |
| increase controller nodes from 2 to 4 | CCP approximate algorithm · Battle Meeting | group objective computation time (sec.): 17.6 -> 322.4 | Hurt |
| increase controller nodes from 4 to 6 | CCP approximate algorithm · Battle Meeting | group objective computation time (sec.): 322.4 -> 181.5 | Improved |
| increase controller nodes from 2 to 6 | CCP approximate algorithm · Prisoner Meeting | individual objective computation time (sec.): 20.7 -> 182.2 | Hurt |
| increase allotted slack from 0 to 25 | CCP approximate algorithm · Battle Meeting | group objective ADR | Hurt |
| increase allotted slack from 0 to 25 | CCP approximate algorithm · Battle Meeting | individual objective ADR | Improved |
| increase allotted slack from 0 to 25 | CCP approximate algorithm · Prisoner Meeting | individual objective ADR | Hurt |
Closest papers by meaning
- Multi-Robot Task Allocation Using Global Games with Negative Feedback: The Colony Maintenance Problem (AAAI 2025 · similarity 0.86)
- Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination (ICML 2025 · similarity 0.86)
- Team-Fictitious Play for Reaching Team-Nash Equilibrium in Multi-team Games (NeurIPS 2024 · similarity 0.85)
- POGEMA: A Benchmark Platform for Cooperative Multi-Agent Pathfinding (ICLR 2025 · similarity 0.85)
- Multiagent Gumbel MuZero: Efficient Planning in Combinatorial Action Spaces (AAAI 2024 · similarity 0.85)
- COSAC: Counterfactual Credit Assignment in Sequential Cooperative Teams (NeurIPS 2026 · similarity 0.85)
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.
