Offline Multi-Task Reinforcement Learning

Multi-Task Distributional RL

Framework modeling the complete distribution of returns rather than their expectation for each task in an offline multi-task context.

← Tillbaka