Single-Agent Policy Tree Search With Guarantees.
Laurent OrseauLevi LelisTor LattimoreTheophane WeberPublished in: NeurIPS (2018)
Keyphrases
- tree search
- single agent
- path finding
- larger problems
- multi agent
- action space
- policy gradient
- search algorithm
- multiple agents
- heuristic search
- decision problems
- dynamic environments
- multi agent systems
- optimal policy
- branch and bound
- path planning
- reinforcement learning
- constraint propagation
- mathematical programming
- search tree
- state space
- iterative deepening
- partially observable markov decision processes
- special case
- lower bound
- cooperative
- concept learning
- search space
- monte carlo
- orders of magnitude
- dynamic programming