A robust policy bootstrapping algorithm for multi-objective reinforcement learning in non-stationary environments.
Sherif M. AbdelfattahKathryn KasmarikJiankun HuPublished in: Adapt. Behav. (2020)
Keyphrases
- multi objective
- learning algorithm
- dynamic programming
- reinforcement learning
- optimization algorithm
- preprocessing
- optimal policy
- computationally efficient
- np hard
- computational complexity
- optimal solution
- objective function
- detection algorithm
- probabilistic model
- evolutionary algorithm
- worst case
- model free
- non stationary
- function approximation
- test problems
- machine learning
- optimization problems
- mobile robot
- active learning
- lower bound
- similarity measure
- clustering algorithm
- genetic algorithm