Provable and Practical: Efficient Exploration in Reinforcement Learning via Langevin Monte Carlo.
Haque IshfaqQingfeng LanPan XuA. Rupam MahmoodDoina PrecupAnima AnandkumarKamyar AzizzadenesheliPublished in: ICLR (2024)
Keyphrases
- monte carlo
- reinforcement learning
- temporal difference
- stochastic approximation
- policy evaluation
- importance sampling
- monte carlo simulation
- markov chain
- monte carlo methods
- variance reduction
- optimal strategy
- function approximation
- adaptive sampling
- reinforcement learning algorithms
- monte carlo tree search
- monte carlo method
- temporal difference learning
- markovian decision
- game tree
- model free
- machine learning
- function approximators
- control problems
- simulation study
- upper bound
- state space