Research map / Multi-agent systems and games / Cooperative multi-agent RL
ALMA: Hierarchical Learning for Composite Multi-Agent Tasks
Shariq Iqbal, Robby Costales, Fei Sha
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 |
|---|---|---|---|
| learned allocation (ALMA) instead of Flat baselines (A-QMIX / REFIL) | ALMA · SaveTheCity and StarCraft | success rate / return | Improved |
| learned allocation instead of COPA hierarchical baseline | ALMA · StarCraft MMM symmetric | success rate | No clear effect |
| learned allocation instead of hand-designed allocation heuristics (ablation) | ALMA · SaveTheCity and StarCraft | success rate | Improved |
| add subtask-specific observation masking (sub-environment decomposition) (ablation) | ALMA · StarCraft and SaveTheCity | success rate | Improved |
| train allocation and action controllers jointly rather than reuse heuristic-trained action policies (ablation) | ALMA · StarCraft S&Z disadvantage | success rate | Improved |
| move enemy armies closer (violating subtask independence) (ablation) | ALMA · StarCraft S&Z disadvantage | success rate | Mixed |
Closest papers by meaning
- Multi-Agent Meta-Reinforcement Learning: Sharper Convergence Rates with Task Similarity (NeurIPS 2023 · similarity 0.89)
- Distilling Task-Level Coordination Policies for Generalizable Multi-Agent Cooperation (ICML 2026 · similarity 0.89)
- LDSA: Learning Dynamic Subtask Assignment in Cooperative Multi-Agent Reinforcement Learning (NeurIPS 2022 · similarity 0.89)
- HyperMARL: Adaptive Hypernetworks for Multi-Agent RL (NeurIPS 2025 · similarity 0.88)
- Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learning (ICLR 2026 · similarity 0.88)
- Learning Generalizable Skills from Offline Multi-Task Data for Multi-Agent Cooperation (ICLR 2025 · 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.
