Offline Multi-Task Reinforcement Learning

Task-Agnostic Representation Learning

Process of learning state-action representations that are generalizable from batch data without specific knowledge of future tasks.

← Zurück