arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.06315cs.CV
收藏

DETA:面向少样本学习的去噪任务适配

DETA: Denoised Task Adaptation for Few-Shot Learning

University of Electronic Science and Technology of China (UESTC)(电子科技大学);Shenzhen Institute for Advanced Study, UESTC(电子科技大学深圳高等研究院)

Ji Zhang, Lianli Gao, Xu Luo, Hengtao Shen, Jingkuan Song

AI总结:

本文提出 DETA,一个无需额外监督、可与现有任务适配方法正交结合的统一图像与标签去噪框架,在 Meta-Dataset 上显著提升多种基线并刷新最先进性能。

摘要:

少样本学习中的测试时任务适配,旨在仅依靠少量带标签的支持样本,使预训练的任务无关模型适配并捕获测试任务的任务特定知识。以往方法通常侧重于开发先进算法来实现这一目标,却忽视了给定支持样本本身固有的问题。事实上,在仅有少量样本可用的情况下,来自支持样本的图像噪声(又称 X-noise)或标签噪声(又称 Y-noise)所产生的不利影响都可能被严重放大。为应对这一挑战,本文提出 DEnoised Task Adaptation(DETA,去噪任务适配),这是首个统一的图像与标签去噪框架,并且与现有任务适配方法相互正交。在无需额外监督的情况下,DETA 同时利用支持样本的全局视觉信息和局部区域细节,过滤与任务无关的含噪表征。在具有挑战性的 Meta-Dataset 上,DETA 持续提升了应用于多种预训练模型的一系列基线方法的性能。值得注意的是,通过处理 Meta-Dataset 中此前被忽视的图像噪声,DETA 取得了新的最先进结果。代码已发布于 https://github.com/JimZAI/DETA。

英文摘要:

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/JimZAI/DETA.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑