MDPs with Energy-Parity Objectives.
Richard MayrSven SchewePatrick TotzkeDominik WojtczakPublished in: CoRR (2017)
Keyphrases
- markov decision processes
- reinforcement learning
- energy consumption
- state space
- error correction
- dynamic programming
- factored mdps
- finite horizon
- multiple objectives
- energy minimization
- dec pomdps
- energy efficiency
- semi markov decision processes
- average reward
- low energy
- decision theoretic planning
- model based reinforcement learning
- markov decision process
- energy saving