Toward negotiable reinforcement learning: shifting priorities in Pareto optimal sequential decision-making.
Andrew CritchPublished in: CoRR (2017)
Keyphrases
- sequential decision making
- pareto optimal
- reinforcement learning
- expected utility
- multi objective
- multi objective optimization
- interactive dynamic influence diagrams
- multiple objectives
- function approximation
- nash equilibrium
- temporal difference
- multi issue negotiation
- state space
- nsga ii
- model free
- reinforcement learning algorithms
- optimal policy
- optimal solution
- markov decision processes
- evolutionary algorithm
- multi agent
- learning algorithm
- pareto optimal set
- decision problems
- action selection
- influence diagrams
- decision making
- supervised learning
- dynamic programming
- special case