Research map / Multi-agent systems and games / Cooperative multi-agent RL
Successor Features Based Multi-Agent RL for Event-Based Decentralized MDPs
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 |
|---|---|---|---|
| Replace GKP deep RL with Dec-ESR actor-critic | GKP · MRT coverage, Easy | solution quality gain: 28% | Helped |
| Replace GKP with Dec-ESR | GKP · MRT coverage, Medium | solution quality gain: 42.8% | Helped |
| Replace GKP with Dec-ESR | GKP · MRT coverage, Hard | solution quality gain: 49.8% | Helped |
| Add event-based critic (Dec-ESR vs No Event Critic) (ablation) | NEC · MRT coverage, Easy | solution quality gain: 18.9% | Helped |
| Add event-based critic (ablation) | NEC · MRT coverage, Medium | solution quality gain: 28.7% | Helped |
| Add event-based critic (ablation) | NEC · MRT coverage, Hard | solution quality gain: 32.5% | Helped |
| Local critic only (NEC) instead of GKP (ablation) | GKP · MRT coverage | solution quality | Helped |
| Transfer learned parameters to new reward task | Dec-ESR · MRT coverage, changed rewards | asymptotic global reward: around 5% higher | Helped |
| Transfer when only joint reward changes | Dec-ESR · MRT coverage | iterations to converge: 3000 -> 700 | Helped |
| Transfer when joint and local rewards change | Dec-ESR · MRT coverage | iterations to converge: 3000 -> 1000 | Helped |
| Scale to synthetic map with up to 30 agents (Easy) | Dec-ESR vs GKP · Synthetic MRT map | solution quality gain: 37.95% | Helped |
| Scale to synthetic map (Hard) | Dec-ESR vs GKP · Synthetic MRT map | solution quality gain: 52% | Helped |
Closest papers by meaning
- Planning and Learning for Decentralized MDPs With Event Driven Rewards (AAAI 2018 · similarity 0.90)
- Finite-Time Global Optimality Convergence in Deep Neural Actor-Critic Methods for Decentralized Multi-Agent Reinforcement Learning (ICML 2025 · similarity 0.86)
- Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems (NeurIPS 2026 · similarity 0.86)
- Federated Natural Policy Gradient and Actor Critic Methods for Multi-task Reinforcement Learning (NeurIPS 2024 · similarity 0.85)
- Heterogeneous Agent Collaborative Reinforcement Learning (NeurIPS 2026 · similarity 0.85)
- Optimally Solving Simultaneous-Move Dec-POMDPs: The Sequential Central Planning Approach (AAAI 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.