Continuous Multi-Objective Reinforcement Learning

Preference-based RL

An approach where human preferences on trade-offs between objectives are integrated into the learning process to guide the agent towards desirable solutions on the Pareto front.

← Zurück