ray.rllib.core.learner.learner_group.LearnerGroup.additional_update#
- LearnerGroup.additional_update(*, reduce_fn=-1, **kwargs) Dict[str, Any] | List[Dict[str, Any]] [source]#
Apply additional non-gradient based updates to the Learners.
For example, this could be used to do a polyak averaging update of a target network in off policy algorithms like SAC or DQN.
By default, this is a pass through that calls all Learner workers’
additional_update(**kwargs)
method.- Returns:
A list of dictionaries of results returned by the
Learner.additional_update()
calls.