On the Convergence of the Monte Carlo Exploring Starts Algorithm for Reinforcement Learning.
Che WangShuhan YuanKai ShaoKeith W. RossPublished in: ICLR (2022)
Keyphrases
- monte carlo
- stochastic approximation
- learning algorithm
- dynamic programming
- importance sampling
- computational complexity
- objective function
- monte carlo tree search
- computational cost
- matrix inversion
- temporal difference learning
- detection algorithm
- convergence rate
- reinforcement learning
- optimal strategy
- monte carlo simulation
- optimal solution
- simulation study
- expectation maximization
- particle filter
- markov chain
- simulated annealing
- temporal difference
- iterative algorithms
- variance reduction
- monte carlo methods