Research map / Multi-agent systems and games / Cooperative multi-agent RL
LIGS: Learnable Intrinsic-Reward Generation Selection for Multi-Agent 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 LIGS learned intrinsic rewards and switching controls | MAPPO agents · Foraging coordinated exploration | performance | Helped |
| use LIGS to guide agents away from low-reward joint policies | Foraging agents · Foraging optimal joint policies | reward acquisition speed: around 400k episodes | Helped |
| add LIGS learned intrinsic rewards and switching controls | MARL agents · Foraging sparse rewards | learning performance | No clear effect |
| add LIGS to IPPO (ablation) | IPPO · Foraging corridor coordination | task performance | Helped |
| replace learned switching controls with random equal-probability switching (ablation) | LIGS · Foraging coordination | performance | Hurt |
| replace learned switching controls with always-on intrinsic rewards (ablation) | LIGS · Foraging coordination | performance | Hurt |
| replace the RND exploration bonus with a count-based bonus (ablation) | LIGS · Foraging tasks | performance | No clear effect |
| replace the RND exploration bonus with a count-based bonus (ablation) | LIGS · Foraging joint exploration | performance | Helped |
| add RND intrinsic rewards directly to the agents' objectives instead of using LIGS reward selection and switching (ablation) | MAPPO · Foraging coordination | performance | Hurt |
| use LIGS rather than the baseline MARL methods | MARL agents · SMAC | win rate | Helped |
Closest papers by meaning
- LIIR: Learning Individual Intrinsic Reward in Multi-Agent Reinforcement Learning (NeurIPS 2019 · similarity 0.90)
- Individual Contributions as Intrinsic Exploration Scaffolds for Multi-agent Reinforcement Learning (ICML 2024 · similarity 0.87)
- First-Explore, then Exploit: Meta-Learning to Solve Hard Exploration-Exploitation Trade-Offs (NeurIPS 2024 · similarity 0.87)
- Foundations of Multi-Agent Learning in Dynamic Environments: Where Reinforcement Learning Meets Strategic Decision-Making (AAAI 2025 · similarity 0.87)
- FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.87)
- Optimistic Multi-Agent Policy Gradient (ICML 2024 · 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.