Memory Bounded Open-Loop Planning in Large POMDPs Using Thompson Sampling.
Thomy PhanLenz BelznerMarie KiermeierMarkus FriedrichKyrill SchmidClaudia Linnhoff-PopienPublished in: AAAI (2019)
Keyphrases
- open loop
- closed loop
- control system
- partially observable markov decision processes
- feedback control
- belief state
- belief space
- partially observable
- reinforcement learning
- planning problems
- control law
- predictive state representations
- monte carlo
- point based value iteration
- control scheme
- markov decision processes
- decision problems
- inverted pendulum
- state space
- dynamic programming
- evolutionary algorithm
- expert systems