Continuous Multi-Objective Reinforcement Learning

Dynamic Weighting

Adaptive strategy that modifies the weights of objectives during learning to efficiently explore the Pareto front and avoid local optima.

← Kembali