Inverse Q-Learning Done Right: Offline Imitation Learning in $Q^π$-Realizable MDPs

OffRLRe-assign community