Research map / Reinforcement learning / Visual and representation RL
A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement 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 |
|---|---|---|---|
| Include the peer learning gradient alongside the current-policy and own-learning gradients | Meta-MAPG · IPD | adaptation performance | Helped |
| Use Meta-MAPG instead of Meta-PG | Meta-MAPG · IPD, in-distribution personas | AUC: 6.13±0.05 -> 13.77±0.25 | Helped |
| Use Meta-MAPG instead of Meta-PG | Meta-MAPG · IPD, out-of-distribution personas | AUC: 7.60±0.07 -> 11.12±0.33 | Helped |
| Remove the peer learning gradient (ablation) | Meta-MAPG · IPD | adaptation performance | Hurt |
| Remove the own learning gradient (ablation) | Meta-MAPG · IPD | adaptation performance | Hurt |
| Use opponent modeling to estimate peer policy parameters during meta-training (ablation) | Meta-MAPG · IPD | adaptation performance | Hurt |
| Increase the number of trajectories between peer policy updates (ablation) | Meta-MAPG · IPD | AUC | Helped |
| Add the peer learning gradient | Meta-PG · 2-Agent HalfCheetah | adaptation performance | No clear effect |
| Use meta-learning that models own and peer learning dynamics instead of REINFORCE | Meta-MAPG · 2-Agent HalfCheetah | adaptation performance | Helped |
| Use Meta-MAPG instead of Meta-PG | Meta-MAPG · RPS, including 3-player and 4-player settings | adaptation performance | Helped |
Closest papers by meaning
- Multi-agent cooperation through learning-aware policy gradients (ICLR 2025 · similarity 0.91)
- Optimistic Multi-Agent Policy Gradient (ICML 2024 · similarity 0.89)
- Learning to Deliberate: Meta-policy Collaboration for Agentic LLMs with Multi-agent Reinforcement Learning (AAAI 2026 · similarity 0.89)
- Robust and Diverse Multi-Agent Learning via Rational Policy Gradient (NeurIPS 2025 · similarity 0.89)
- Towards Principled Unsupervised Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- HyperMARL: Adaptive Hypernetworks for Multi-Agent RL (NeurIPS 2025 · similarity 0.88)
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.