Pareto Envelope Augmented with Reinforcement Learning: Multi-objective reinforcement learning-based approach for Large-Scale Constrained Pressurized Water Reactor optimization.
Paul SeurinKoroush ShirvanPublished in: CoRR (2023)
Keyphrases
- multi objective
- reinforcement learning
- optimization algorithm
- multiple objectives
- multiobjective optimization
- evolutionary algorithm
- multi objective optimization
- function approximation
- evolutionary optimization
- state space
- reinforcement learning algorithms
- model free
- conflicting objectives
- optimum design
- particle swarm optimization
- genetic algorithm
- objective function
- markov decision processes
- pareto optimal
- learning algorithm
- optimization problems
- optimal policy
- learning process
- trade off
- multi agent
- temporal difference
- real world
- dynamic programming
- estimation of distribution algorithms
- machine learning
- concave convex procedure
- bi objective
- action selection
- global optimization
- optimization method
- fitness function
- supervised learning