Research map / Reinforcement learning / MDPs and planning
Planning and Learning for Decentralized MDPs With Event Driven Rewards
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 |
|---|---|---|---|
| solve independent program per agent with EM instead of joint NLP | TIDec-MDP solver · Mars rover Hard | runtime | Helped |
| use EM instead of NLP | TIDec-MDP solver · Mars rover Hard | instances solved in 30 min: 40% -> always | Helped |
| use EM instead of NLP | TIDec-MDP solver · Mars rover Hard | reward ratio NLP/EM: 71% | Helped |
| use EM instead of NLP | TIDec-MDP solver · Mars rover Easy | reward ratio NLP/EM: 100% | No clear effect |
| run EM for more iterations (anytime) | EM · Mars rover Hard | fraction of final quality: 98.7% -> 99.6% | Helped |
| use MARL instead of EM on small MRT instances | MARL · MRT coverage small | solution quality: 97% | No clear effect |
| use MARL instead of EM on small MRT instances | MARL · MRT coverage small | runtime: 10 minutes -> 5 hours | Hurt |
| coordinated MARL instead of independent policy optimization I-RL | MARL · Singapore MRT, 10 shared locations | quality gain: +36.1%, 29.6%, 26.0%, 15.2% | Helped |
| lower reset time k | MARL vs I-RL · Singapore MRT | gain of MARL over I-RL | Helped |
| lower reset time k | MARL · Singapore MRT | average reward | Helped |
Closest papers by meaning
- Successor Features Based Multi-Agent RL for Event-Based Decentralized MDPs (AAAI 2019 · similarity 0.90)
- Collective Multiagent Sequential Decision Making Under Uncertainty (AAAI 2017 · similarity 0.88)
- Policy Gradient With Value Function Approximation For Collective Multiagent Planning (NeurIPS 2017 · similarity 0.88)
- Learning to Act in Decentralized Partially Observable MDPs (ICML 2018 · similarity 0.87)
- Learning for Decentralized Control of Multiagent Systems in Large, Partially-Observable Stochastic Environments (AAAI 2016 · similarity 0.87)
- Regret Lower Bounds for Decentralized Multi-Agent Stochastic Shortest Path Problems (NeurIPS 2025 · 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.