Model-Based Multi-objective Reinforcement Learning with Unknown Weights.
Tomohiro YamaguchiShota NagahamaYoshihiro IchikawaKeiki TakadamaPublished in: HCI (5) (2019)
Keyphrases
- multi objective
- reinforcement learning
- model free
- evolutionary algorithm
- multi objective optimization
- optimization algorithm
- function approximation
- objective function
- genetic algorithm
- multiple objectives
- weighted sum
- linear combination
- conflicting objectives
- multiobjective optimization
- state space
- multi objective optimization problems
- particle swarm optimization
- data driven
- relative importance
- bi objective
- machine learning
- nsga ii
- temporal difference
- optimal control
- initially unknown
- reinforcement learning algorithms
- weighting scheme
- markov decision processes
- trade off
- multi criteria
- weighted average
- learning problems
- mobile robot
- multiobjective evolutionary algorithm
- multi objective genetic algorithms