Learning to Defer to a Population: A Meta-Learning Approach

arXiv:2403.02683v1 Announce Type: new
Abstract: The learning to defer (L2D) framework allows autonomous systems to be safe and robust by allocating difficult decisions to a human expert. All existing work on L2D assumes that each expert is well-identified, and if any expert were to change, the system should be re-trained. In this work, we alleviate this constraint, formulating an L2D system that can cope with never-before-seen experts at test-time. We accomplish this by using meta-learning, considering both optimization- and model-based …

abstract arxiv autonomous autonomous systems change cs.lg decisions expert framework human meta meta-learning population robust stat.ml systems type work