Meta Automatic Curriculum Learning for Classrooms of Deep RL Black-Box Students
Abstract: A major challenge in the Deep RL (DRL) community is to train agents able to generalize their control policy over situations never seen in training. Training on diverse tasks has been identified as a key ingredient for good generalization, which pushed researchers towards using rich procedural task generation systems controlled through complex continuous parameter spaces. In such complex continuous task spaces, it is essential to rely on some form of Automatic Curriculum Learning (ACL) to adapt the task sampling distribution to a given learning agent, instead of randomly sampling tasks, as many could end up being either trivial or unfeasible. Since it is hard to get prior knowledge on such task spaces, many ACL algorithms explore the task space to detect progress niches over time. This costly tabula rasa search process needs to be performed for each new learning agents, although they might have similarities in their capabilities profiles. To address this limitation, we introduce the concept of Meta-ACL, and formalize it in the context of black-box RL learners, i.e. algorithms seeking to generalize curriculum generation to an (unknown) distribution of learners. In this work, we present AGAIN, a first instantiation of Meta-ACL, and showcase its benefits for curriculum generation over classical ACL in multiple simulated environments including procedurally generated parkour environments with learners of varying morphologies. Videos and code are available at https://sites.google.com/view/meta-acl.
Submission Length: Long submission (more than 12 pages of main content)
Changes Since Last Submission: Changes performed to the manuscript are highlighted in red.
All modifications are presented in answers to reviewers.
Modifications are mainly focused on the formalization (section 3) and the presentation of our method (section 4).
Assigned Action Editor: ~Olivier_Pietquin1
Submission Number: 399
Loading