PA2D-MORL: Pareto Ascent Directional Decomposition Based Multi-Objective Reinforcement Learning.
Tianmeng HuBiao LuoPublished in: AAAI (2024)
Keyphrases
- multi objective
- reinforcement learning
- multi objective optimization
- evolutionary algorithm
- optimization algorithm
- multiobjective optimization
- function approximation
- objective function
- multiple objectives
- steerable pyramid
- particle swarm optimization
- multi objective optimization problems
- genetic algorithm
- evolutionary optimization
- bi objective
- pareto optimal
- state space
- nsga ii
- pareto optimal solutions
- decomposition method
- conflicting objectives
- learning algorithm
- model free
- temporal difference
- reinforcement learning algorithms
- trade off
- machine learning
- robotic control
- multiobjective evolutionary algorithm
- multiagent evolutionary algorithm
- optimal control
- pareto set
- pareto dominance
- search space
- discrete wavelet transform
- learning process