Reinforcement Learning of Risk-Constrained Policies in Markov Decision Processes.
Tomás BrázdilKrishnendu ChatterjeePetr NovotnýJiri VahalaPublished in: CoRR (2020)
Keyphrases
- markov decision processes
- optimal policy
- reinforcement learning
- risk sensitive
- markov decision process
- reward function
- state space
- reinforcement learning algorithms
- average cost
- policy iteration
- finite state
- policy iteration algorithm
- decision processes
- control policies
- macro actions
- infinite horizon
- decentralized control
- total reward
- decision problems
- finite horizon
- action space
- dynamic programming
- expected reward
- model based reinforcement learning
- state and action spaces
- partially observable
- partially observable markov decision processes
- average reward
- multistage
- markov decision problems
- model free
- function approximation
- factored mdps
- decision theoretic planning
- discounted reward
- decision making
- transition matrices
- planning under uncertainty
- stationary policies
- reachability analysis
- long run
- stochastic games
- state abstraction
- multiagent reinforcement learning
- optimal control
- action sets
- control policy
- hierarchical reinforcement learning
- continuous state spaces
- policy gradient
- machine learning
- temporal difference
- learning algorithm