Saved in:
Bibliographic Details
Main Authors: Lee, Jae-Jun, Yoon, Sung Whan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.06768
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914709451571200
author Lee, Jae-Jun
Yoon, Sung Whan
author_facet Lee, Jae-Jun
Yoon, Sung Whan
contents 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, where they are linearly combined to form an effective initialization to a given task. XB-MAML observes the discrepancy between the vector space spanned by the basis and fine-tuned parameters to decide whether to expand the basis. Our method surpasses the existing works in the multi-domain meta-learning benchmarks and opens up new chances of meta-learning for obtaining the diverse inductive bias that can be combined to stretch toward the effective initialization for diverse unseen tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2403_06768
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle XB-MAML: Learning Expandable Basis Parameters for Effective Meta-Learning with Wide Task Coverage
Lee, Jae-Jun
Yoon, Sung Whan
Machine Learning
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, where they are linearly combined to form an effective initialization to a given task. XB-MAML observes the discrepancy between the vector space spanned by the basis and fine-tuned parameters to decide whether to expand the basis. Our method surpasses the existing works in the multi-domain meta-learning benchmarks and opens up new chances of meta-learning for obtaining the diverse inductive bias that can be combined to stretch toward the effective initialization for diverse unseen tasks.
title XB-MAML: Learning Expandable Basis Parameters for Effective Meta-Learning with Wide Task Coverage
topic Machine Learning
url https://arxiv.org/abs/2403.06768