Assisting elders via dynamic multi-tasks planning: a Markov decision processes based approach.
François CourtemancheMehdi NajjarBlandine PaccoudAndré MayersPublished in: AMBI-SYS (2008)
Keyphrases
- markov decision processes
- decision theoretic planning
- macro actions
- state space
- partially observable
- planning under uncertainty
- optimal policy
- reinforcement learning
- finite state
- reachability analysis
- dynamic programming
- finite horizon
- policy iteration
- transition matrices
- decision processes
- model based reinforcement learning
- probabilistic planning
- average cost
- dynamic environments
- factored mdps
- action space
- heuristic search
- reinforcement learning algorithms
- partially observable markov decision processes
- planning problems
- markov decision problems
- long run
- optimal control
- average reward
- stationary policies
- dynamical systems
- infinite horizon
- state and action spaces
- initial state
- machine learning
- search algorithm
- learning algorithm