V-trace

E1276522 UNEXPLORED

V-trace is an off-policy actor-critic algorithm for deep reinforcement learning that corrects for discrepancies between behavior and target policies using truncated importance sampling.

All labels observed (1)

Label Occurrences
V-trace canonical 2

How this entity was disambiguated

Referenced by (2)

Full triples — surface form annotated when it differs from this entity's canonical label.

IMPALA uses V-trace
IMPALA hasAlgorithm V-trace