IntervalMDP.jl: Accelerated Value Iteration for Interval Markov Decision Processes.
Frederik Baymler MathiesenMorteza LahijanianLuca LaurentiPublished in: CoRR (2024)
Keyphrases
- markov decision processes
- state space
- finite state
- optimal policy
- policy iteration
- reinforcement learning
- transition matrices
- dynamic programming
- planning under uncertainty
- risk sensitive
- partially observable
- average reward
- markov decision process
- decision processes
- infinite horizon
- semi markov decision processes
- action sets
- model based reinforcement learning
- decision theoretic planning
- finite horizon
- reinforcement learning algorithms
- average cost
- factored mdps
- reachability analysis
- stochastic shortest path
- data mining
- partially observable markov decision processes
- discounted reward
- discount factor
- reward function