Research map / Multi-agent systems and games / Games and equilibria
Evaluating and Rewarding Teamwork Using Cooperative Game Abstractions
Tom Yan, Christian Kroer, Alexander Peysakhovich
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 |
|---|---|---|---|
| Model pairwise interactions with a second-order CGA rather than summing player contributions with a first-order CGA (ablation) | CGA · Virtual Teams test set | MSE: $.79$ -> $.07$ | Improved |
| Use DeepSet instead of a second-order CGA | Virtual Teams · Virtual Teams test set | MSE: $.07$ -> $.042$ | Improved |
| Model pairwise interactions with a second-order CGA rather than using a first-order CGA (ablation) | CGA · Virtual Teams best-team formation | score ratio to actual best team | Improved |
| Use a second-order CGA instead of DeepSet for team formation | Virtual Teams · Virtual Teams best-team formation | score ratio to actual best team | Improved |
| Model pairwise interactions with a second-order CGA rather than using a first-order CGA (ablation) | CGA · NBA test set | mean NLL: $-.631$ -> $-.627$ | Improved |
| Increase CGA order from second to third (ablation) | CGA · NBA test set | mean NLL | No clear effect |
| Replace second-order CGA with DeepSet | NBA team performance prediction · NBA test set | mean NLL: $-.627$ -> $-.63.$ | No clear effect |
Closest papers by meaning
- CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement (ICML 2026 · similarity 0.81)
- Ad Hoc Teamwork via Offline Goal-Based Decision Transformers (ICML 2025 · similarity 0.80)
- Open Ad Hoc Teamwork with Cooperative Game Theory (ICML 2024 · similarity 0.80)
- Enhancing Human Experience in Human-Agent Collaboration: A Human-Centered Modeling Approach Based on Positive Human Gain (ICLR 2024 · similarity 0.80)
- Team-Fictitious Play for Reaching Team-Nash Equilibrium in Multi-team Games (NeurIPS 2024 · similarity 0.79)
- Aligning Individual and Collective Objectives in Multi-Agent Cooperation (NeurIPS 2024 · similarity 0.78)
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.
