RDA: Robust Domain Adaptation via Fourier Adversarial Attacking

06/05/2021
by   Jiaxing Huang, et al.
0

Unsupervised domain adaptation (UDA) involves a supervised loss in a labeled source domain and an unsupervised loss in an unlabeled target domain, which often faces more severe overfitting (than classical supervised learning) as the supervised source loss has clear domain gap and the unsupervised target loss is often noisy due to the lack of annotations. This paper presents RDA, a robust domain adaptation technique that introduces adversarial attacking to mitigate overfitting in UDA. We achieve robust domain adaptation by a novel Fourier adversarial attacking (FAA) method that allows large magnitude of perturbation noises but has minimal modification of image semantics, the former is critical to the effectiveness of its generated adversarial samples due to the existence of 'domain gaps'. Specifically, FAA decomposes images into multiple frequency components (FCs) and generates adversarial samples by just perturbating certain FCs that capture little semantic information. With FAA-generated samples, the training can continue the 'random walk' and drift into an area with a flat loss landscape, leading to more robust domain adaptation. Extensive experiments over multiple domain adaptation tasks show that RDA can work with different computer vision tasks with superior performance.

READ FULL TEXT
research
11/18/2020

Robustified Domain Adaptation

Unsupervised domain adaptation (UDA) is widely used to transfer a model ...
research
09/02/2021

Adversarial Robustness for Unsupervised Domain Adaptation

Extensive Unsupervised Domain Adaptation (UDA) studies have shown great ...
research
02/18/2022

Exploring Adversarially Robust Training for Unsupervised Domain Adaptation

Unsupervised Domain Adaptation (UDA) methods aim to transfer knowledge f...
research
06/16/2020

Mining Label Distribution Drift in Unsupervised Domain Adaptation

Unsupervised domain adaptation targets to transfer task knowledge from l...
research
02/09/2018

Unsupervised Deep Domain Adaptation for Pedestrian Detection

This paper addresses the problem of unsupervised domain adaptation on th...
research
03/08/2022

Robust Local Preserving and Global Aligning Network for Adversarial Domain Adaptation

Unsupervised domain adaptation (UDA) requires source domain samples with...
research
01/13/2021

Adversarial Sample Enhanced Domain Adaptation: A Case Study on Predictive Modeling with Electronic Health Records

With the successful adoption of machine learning on electronic health re...

Please sign up or login with your details

Forgot password? Click here to reset