XB-MAML: Learning Expandable Basis Parameters for Effective Meta-Learning with Wide Task Coverage

arXiv:2403.06768v1 Announce Type: new
Abstract: Meta-learning, which pursues an effective initialization model, has emerged as a promising approach to handling unseen tasks. However, a limitation remains to be evident when a meta-learner tries to encompass a wide range of task distribution, e.g., learning across distinctive datasets or domains. Recently, a group of works has attempted to employ multiple model initializations to cover widely-ranging tasks, but they are limited in adaptively expanding initializations. We introduce XB-MAML, which learns expandable basis parameters, …

abstract arxiv coverage cs.lg datasets distribution domains however meta meta-learning parameters tasks type