DETA: Denoised Task Adaptation for Few-Shot Learning

03/11/2023
by   Ji Zhang, et al.
0

Test-time task adaptation in few-shot learning aims to adapt a pre-trained task-agnostic model for capturing taskspecific knowledge of the test task, rely only on few-labeled support samples. Previous approaches generally focus on developing advanced algorithms to achieve the goal, while neglecting the inherent problems of the given support samples. In fact, with only a handful of samples available, the adverse effect of either the image noise (a.k.a. X-noise) or the label noise (a.k.a. Y-noise) from support samples can be severely amplified. To address this challenge, in this work we propose DEnoised Task Adaptation (DETA), a first, unified image- and label-denoising framework orthogonal to existing task adaptation approaches. Without extra supervision, DETA filters out task-irrelevant, noisy representations by taking advantage of both global visual information and local region details of support samples. On the challenging Meta-Dataset, DETA consistently improves the performance of a broad spectrum of baseline methods applied on various pre-trained models. Notably, by tackling the overlooked image noise in Meta-Dataset, DETA establishes new state-of-the-art results. Code is released at https://github.com/nobody-1617/DETA.

READ FULL TEXT

page 1

page 3

page 8

research
07/01/2021

Improving Task Adaptation for Cross-domain Few-shot Learning

In this paper, we look at the problem of cross-domain few-shot classific...
research
09/28/2020

Interventional Few-Shot Learning

We uncover an ever-overlooked deficiency in the prevailing Few-Shot Lear...
research
03/20/2023

Feature Alignment and Uniformity for Test Time Adaptation

Test time adaptation (TTA) aims to adapt deep neural networks when recei...
research
12/31/2022

Unlearnable Clusters: Towards Label-agnostic Unlearnable Examples

There is a growing interest in developing unlearnable examples (UEs) aga...
research
04/12/2022

Few-shot Learning with Noisy Labels

Few-shot learning (FSL) methods typically assume clean support sets with...
research
10/10/2022

Multi-Modal Fusion by Meta-Initialization

When experience is scarce, models may have insufficient information to a...
research
05/25/2021

Bridging Few-Shot Learning and Adaptation: New Challenges of Support-Query Shift

Few-Shot Learning (FSL) algorithms have made substantial progress in lea...

Please sign up or login with your details

Forgot password? Click here to reset