A myopic Monte Carlo strategy for the partially observable travelling salesman problem.
Andrew R. BuckJames M. KellerPublished in: CEC (2016)
Keyphrases
- monte carlo
- partially observable
- infinite horizon
- optimal strategy
- decision problems
- state space
- markov decision processes
- reinforcement learning
- markov chain
- monte carlo simulation
- dynamical systems
- importance sampling
- quasi monte carlo
- monte carlo tree search
- markov decision problems
- particle filter
- partial observations
- optimal control
- least squares
- variance reduction
- temporal difference
- reward function
- belief state
- machine learning
- optimal policy
- computational complexity
- multi agent
- transition probabilities
- search strategies
- orders of magnitude
- heuristic search