Value Iteration in Continuous Actions, States and Time.
Michael LutterShie MannorJan PetersDieter FoxAnimesh GargPublished in: ICML (2021)
Keyphrases
- action space
- markov decision processes
- initial state
- state space
- state transitions
- belief state
- perceptual aliasing
- continuous action
- action sequences
- partial knowledge
- partially observable
- state action
- state transition
- belief space
- goal state
- decision theoretic
- state information
- situation calculus
- heuristic search
- dynamic programming
- markov decision process
- internal states
- partially observable markov decision processes
- goal directed
- optimal policy
- decision processes
- markov decision problems
- plan recognition
- markov decision chains
- reasoning about actions
- policy iteration
- reinforcement learning
- cognitive states
- action selection
- state variables
- human actions