Offline Imitation Learning

Transition set

Data structure storing tuples (state, action, next state, reward) extracted from expert trajectories for offline training.

← Indietro