MDPs with energy-parity objectives.
Richard MayrSven SchewePatrick TotzkeDominik WojtczakPublished in: LICS (2017)
Keyphrases
- markov decision processes
- reinforcement learning
- energy consumption
- energy minimization
- state space
- markov decision problems
- neural network
- planning under uncertainty
- factored mdps
- optimal policy
- partially observable
- error correction
- finite horizon
- markov decision process
- average cost
- energy saving
- decision theoretic
- dynamic programming
- machine learning