Decentralized Monte Carlo Tree Search for Partially Observable Multi-Agent Pathfinding.
Alexey SkrynnikAnton AndreychukKonstantin S. YakovlevAleksandr PanovPublished in: AAAI (2024)
Keyphrases
- partially observable
- path finding
- monte carlo tree search
- multi agent
- reinforcement learning
- monte carlo
- state space
- temporal difference
- heuristic search
- single agent
- decision problems
- markov decision processes
- dynamical systems
- path planning
- search algorithm
- evaluation function
- infinite horizon
- markov decision problems
- tree search
- game tree
- multi agent systems
- multiple agents
- markov chain
- optimal path
- function approximation
- belief state
- reward function
- reinforcement learning algorithms
- partially observable markov decision processes
- learning algorithm
- rule learning
- machine learning
- search space
- long run
- hill climbing
- search strategies
- policy iteration