Research map / Multi-agent systems and games / Games and equilibria
Lenient Learning in Independent-Learner Stochastic Cooperative Games
Ermo Wei, Sean Luke
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 |
|---|---|---|---|
| replace Boltzmann selection with Ranked Boltzmann (w=10) (ablation) | LMRL2 · Gradient 1 | complete-policy count: 8407 -> 9983 | Improved |
| replace Boltzmann selection with Ranked Boltzmann (w=10) (ablation) | LMRL2 · RO 3 | complete-policy count: 7332 -> 1739 | Hurt |
| replace Boltzmann selection with Ranked Boltzmann (w=10) (ablation) | LMRL2 · RO 2 | complete-policy count: 9999 -> 5267 | Hurt |
| replace Boltzmann selection with Ranked Boltzmann (w=10) (ablation) | LMRL2 · Climb-FS | complete-policy count: 9016 -> 9217 | Improved |
| add latching (temperature only heats up earlier states) (ablation) | LMRL2 · Gradient 1 | complete-policy count: 8407 -> 7851 | Hurt |
| add latching (temperature only heats up earlier states) (ablation) | LMRL2 · RO 3 | complete-policy count: 7332 -> 767 | Hurt |
| add latching (temperature only heats up earlier states) (ablation) | LMRL2 · Heaven and Hell | correct-policy count: 9991 -> 10000 | Improved |
| pair LMRL2 with Distributed Q-Learning instead of self-play | LMRL2 · RO 1 | complete-policy count: 9368 -> 9998 | Improved |
| pair LMRL2 with Q-Learning instead of self-play | LMRL2 · RO 2 | complete-policy count: 9999 -> 0 | Hurt |
| pair LMRL2 with Q-Learning instead of self-play | LMRL2 · Climb | complete-policy count: 9999 -> 27 | Hurt |
| lower learning rate to alpha=0.05 | LMRL2 · Gradient 1 | complete-policy count: 8407 -> 9207 | Improved |
| lower learning rate to alpha=0.05 | LMRL2 · RO 1 | complete-policy count: 9368 -> 9750 | Improved |
Closest papers by meaning
- Foundations of Multi-Agent Learning in Dynamic Environments: Where Reinforcement Learning Meets Strategic Decision-Making (AAAI 2025 · similarity 0.88)
- Multi-agent cooperation through learning-aware policy gradients (ICLR 2025 · similarity 0.88)
- Incentivize without Bonus: Provably Efficient Model-based Online Multi-agent RL for Markov Games (ICML 2025 · similarity 0.88)
- Multi-Agent Learning under Uncertainty: Recurrence vs. Concentration (NeurIPS 2025 · similarity 0.88)
- Self-Play $Q$-Learners Can Provably Collude in the Iterated Prisoner's Dilemma (ICML 2025 · similarity 0.87)
- Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems (NeurIPS 2026 · 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.
