Research map / Multi-agent systems and games / Cooperative multi-agent RL
SDE-HARL: Scalable Distributed Policy Execution for Heterogeneous-Agent Reinforcement Learning
Toan D. Gian, Mohammad Abdi, Nathaniel D. Bastian, Francesco Restuccia
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 |
|---|---|---|---|
| add parameter sharing (naive) to HARL (ablation) | SDE-HARL · GRF (2 agents) | reward: 1.18 -> 1.24 | Improved |
| add parameter sharing (naive) to HARL (ablation) | SDE-HARL · SMAC/SMACv2 (27 agents) | reward: 11.85 -> 10.65 | Hurt |
| add IAD on top of RAD (ablation) | SDE-HARL · SMAC/SMACv2 (5 agents) | reward: 13.62 -> 16.89 | Improved |
| add RAD on top of IAD (ablation) | SDE-HARL · SMAC/SMACv2 (5 agents) | reward: 15.75 -> 16.89 | Improved |
| add RAD on top of IAD (ablation) | SDE-HARL · SMAC/SMACv2 (27 agents) | reward: 15.64 -> 16.72 | Improved |
| use RAD+IAD instead of contrastive representation learning (ablation) | SDE-HARL · GRF (2 agents) | reward: 1.48 -> 1.58 | Improved |
| use RAD+IAD instead of contrastive representation learning (ablation) | SDE-HARL · SMAC/SMACv2 (27 agents) | reward: 16.77 vs 16.72 | No clear effect |
| add action-aware diversity reward (ablation) | SDE-HARL · SMAC/SMACv2 (5 agents) | reward: 15.67 -> 16.89 | Improved |
| use local split at n=3 blocks instead of n=0 or n=10 (ablation) | SDE-HARL · GRF Run_Pass_and_Shoot (latency/energy) | reward: 1.5x lower inference time vs n=0; 2x vs n=10 | Improved |
| increase lambda to 1e-2 (stronger compression) (ablation) | SDE-HARL · SMACv2 Corridor | reward | Hurt |
Closest papers by meaning
- Heterogeneous Agent Collaborative Reinforcement Learning (NeurIPS 2026 · similarity 0.86)
- Enhancing Cooperative Multi-Agent Reinforcement Learning with State Modelling and Adversarial Exploration (ICML 2025 · similarity 0.85)
- DistRL: An Asynchronous Distributed Reinforcement Learning Framework for On-Device Control Agent (ICLR 2025 · similarity 0.85)
- Cheaper and Faster: Distributed Deep Reinforcement Learning with Serverless Computing (AAAI 2024 · similarity 0.85)
- Learning to Share: Selective Memory for Efficient Parallel Agentic Systems (ICML 2026 · similarity 0.85)
- HyperMARL: Adaptive Hypernetworks for Multi-Agent RL (NeurIPS 2025 · similarity 0.85)
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.
