Continuous Multi-Objective Reinforcement Learning

Non-dominated Solutions

Set of solutions where none is strictly better than another on all objectives, constituting the set of Pareto-optimal solutions.

← Wstecz