Task-Specific Preconditioner for Cross-Domain Few-Shot Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kang, Suhyun, Park, Jungwon, Lee, Wonseok, Rhee, Wonjong
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912162842148864
author Kang, Suhyun
Park, Jungwon
Lee, Wonseok
Rhee, Wonjong
author_facet Kang, Suhyun
Park, Jungwon
Lee, Wonseok
Rhee, Wonjong
contents Cross-Domain Few-Shot Learning~(CDFSL) methods typically parameterize models with task-agnostic and task-specific parameters. To adapt task-specific parameters, recent approaches have utilized fixed optimization strategies, despite their potential sub-optimality across varying domains or target tasks. To address this issue, we propose a novel adaptation mechanism called Task-Specific Preconditioned gradient descent~(TSP). Our method first meta-learns Domain-Specific Preconditioners~(DSPs) that capture the characteristics of each meta-training domain, which are then linearly combined using task-coefficients to form the Task-Specific Preconditioner. The preconditioner is applied to gradient descent, making the optimization adaptive to the target task. We constrain our preconditioners to be positive definite, guiding the preconditioned gradient toward the direction of steepest descent. Empirical evaluations on the Meta-Dataset show that TSP achieves state-of-the-art performance across diverse experimental scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15483
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Task-Specific Preconditioner for Cross-Domain Few-Shot Learning
Kang, Suhyun
Park, Jungwon
Lee, Wonseok
Rhee, Wonjong
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Cross-Domain Few-Shot Learning~(CDFSL) methods typically parameterize models with task-agnostic and task-specific parameters. To adapt task-specific parameters, recent approaches have utilized fixed optimization strategies, despite their potential sub-optimality across varying domains or target tasks. To address this issue, we propose a novel adaptation mechanism called Task-Specific Preconditioned gradient descent~(TSP). Our method first meta-learns Domain-Specific Preconditioners~(DSPs) that capture the characteristics of each meta-training domain, which are then linearly combined using task-coefficients to form the Task-Specific Preconditioner. The preconditioner is applied to gradient descent, making the optimization adaptive to the target task. We constrain our preconditioners to be positive definite, guiding the preconditioned gradient toward the direction of steepest descent. Empirical evaluations on the Meta-Dataset show that TSP achieves state-of-the-art performance across diverse experimental scenarios.
title Task-Specific Preconditioner for Cross-Domain Few-Shot Learning
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.15483