paper-with-me

홈 › Papers

How Does Frequency Bias Affect the Robustness of Neural Image Classifiers against Common Corruption and Adversarial Perturbations?

2022-05-09 · Alvin Chan, Yew-Soon Ong, Clement Tan

Model robustness is vital for the reliable deployment of machine learning models in real-world applications. Recent studies have shown that data augmentation can result in model over-relying on features in the low-frequency domain, sacrificing performance against low-frequency corruptions, highlighting a connection between frequency and robustness. Here, we take one step further to more directly study the frequency bias of a model through the lens of its Jacobians and its implication to model robustness. To achieve this, we propose Jacobian frequency regularization for models' Jacobians to have a larger ratio of low-frequency components. Through experiments on four image datasets, we show that biasing classifiers towards low (high)-frequency components can bring performance gain against high (low)-frequency corruption and adversarial perturbation, albeit with a tradeoff in performance for low (high)-frequency corruption. Our approach elucidates a more direct connection between the frequency bias and robustness of deep learning models.

📄 PDF Abstract BibTeX arXiv:2205.04533

Code (0)

등록된 구현이 없습니다.

Tasks

Data Augmentation

Similar Papers 제목 키워드 기반

Investigating and Explaining the Frequency Bias in Image Classification

2022-05-06 · Zhiyu Lin, YiFei Gao, Jitao Sang

CNNs exhibit many behaviors different from humans, one of which is the capability of employing high-frequency components. This paper discusses the frequency bias phenomenon in image classification tasks: the high-frequen…

Classificationimage-classificationImage Classification

Spatial Frequency Bias in Convolutional Generative Adversarial Networks

2020-10-04 · Mahyar Khayatkhoei, Ahmed Elgammal

As the success of Generative Adversarial Networks (GANs) on natural images quickly propels them into various real-life applications across different domains, it becomes more and more important to clearly understand their…

DenoisingSuper-Resolution

Spatial-frequency channels, shape bias, and adversarial robustness

2023-09-21 · NeurIPS 2023 11

What spatial frequency information do humans and neural networks use to recognize objects? In neuroscience, critical band masking is an established tool that can reveal the frequency-selective filters used for object rec…

The Undesirable Dependence on Frequency of Gender Bias Metrics Based on Word Embeddings

2023-01-02 · Francisco Valentini, Germán Rosati, Diego Fernandez Slezak, Edgar Altszyler

Numerous works use word embedding-based metrics to quantify societal biases and stereotypes in texts. Recent studies have found that word embeddings can capture semantic similarity but may be affected by word frequency. …

Semantic SimilaritySemantic Textual SimilarityWord Embeddings

Dissociating spatial frequency reliance from adversarial robustness advantages in neurally guided deep convolutional neural networks

2026-05-06 · Zhenan Shao, Tianyu Ren, Chengxiao Wang, Leyla Isik 외 arxiv

Deep convolutional neural networks (DCNNs) have rivaled humans on many visual tasks, yet they remain vulnerable to near-imperceptible perturbations generated by adversarial attacks. Recent work shows that aligning DCNN r…

Adversarial RobustnessObject Recognition