{"type":"ScholarlyArticle","title":"Investigating and Explaining the Frequency Bias in Image Classification","authors":[{"name":"Zhiyu Lin","slug":"zhiyu-lin","given":"Zhiyu","family":"Lin","affiliation":"Beijing Jiaotong University , China"},{"name":"Yifei Gao","slug":"yifei-gao","given":"Yifei","family":"Gao","affiliation":"Beijing Jiaotong University , China"},{"name":"Jitao Sang","slug":"jitao-sang","given":"Jitao","family":"Sang","affiliation":"Beijing Jiaotong University , China; Peng Cheng Lab , Shenzhen 518066 , China"}],"abstract":"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-frequency components are actually much less exploited than the low- and mid- frequency components. We first investigate the frequency bias phenomenon by presenting two observations on feature discrimination and learning priority. Furthermore, we hypothesize that (1) the spectral density, (2) class consistency directly affect the frequency bias. Specifically, our investigations verify that the spectral density of datasets mainly affects the learning priority, while the class consistency mainly affects the feature discrimination.","doi":"10.24963/ijcai.2022/101","url":"https://ijcai.org/proceedings/2022/101","pdf":"https://ijcai.org/proceedings/2022/0101.pdf","markdown":"https://ijcai.org/proceedings/2022/101.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/101","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"AI Ethics, Trust, Fairness","track":"Main Track","number":101,"pages":{"first":717,"last":723},"keywords":["AI Ethics, Trust, Fairness: Explainability and Interpretability","Computer Vision: Bias, Fairness & Privacy","AI Ethics, Trust, Fairness: Bias","AI Ethics, Trust, Fairness: Safety & Robustness","AI Ethics, Trust, Fairness: Trustworthy AI"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38987087","length_sec":88,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38985230","length_sec":646,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Abello et al., 2021] Antonio A Abello, Roberto Hirata, and Zhangyang Wang. Dissecting the high-frequency bias in convolutional neural networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 863-871, June 2021.","title":"Dissecting the high-frequency bias in convolutional neural networks","authors":"Antonio A Abello, Roberto Hirata, Zhangyang Wang","year":"2021","venue":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Deng et al., 2009] Jia Deng, Wei Dong, Richard Socher, Li- Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248-255.","title":"Imagenet: A large-scale hierarchical image database","authors":"Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, Li Fei-Fei","year":"2009","venue":"IEEE conference on computer vision and pattern recognition","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Durall et al., 2020] Ricard Durall, Margret Keuper, and Ja- nis Keuper. Watch your up-convolution: Cnn based gener- ative deep neural networks are failing to reproduce spectral distributions. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 7890- 7899, 2020.","title":"Watch your up-convolution: Cnn based generative deep neural networks are failing to reproduce spectral distributions","authors":"Ricard Durall, Margret Keuper, Janis Keuper","year":"2020","venue":"Proceedings of the IEEE/CVF conference on computer vision and pattern recognition","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Dzanic et al., 2020] Tarik Dzanic, Karan Shah, and Freddie Witherden. Fourier spectrum discrepancies in deep net- work generated images. Advances in neural information processing systems, 33:3022-3032, 2020.","title":"Fourier spectrum discrepancies in deep network generated images","authors":"Tarik Dzanic, Karan Shah, Freddie Witherden","year":"2020","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Geirhos et al., 2019] Robert Geirhos, Patricia Rubisch, Claudio Michaelis, Matthias Bethge, Felix A. Wichmann, and Wieland Brendel. Imagenet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. In International Conference on Learning Representations, 2019.","title":"Imagenet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness","authors":"Robert Geirhos, Patricia Rubisch, Claudio Michaelis, Matthias Bethge, Felix A Wichmann, Wieland Brendel","year":"2019","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[He et al., 2016] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recog- nition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770-778, 2016.","title":"Deep residual learning for image recognition","authors":"Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun","year":"2016","venue":"Proceedings of the IEEE conference on computer vision and pattern recognition","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Hermann et al., 2020] Katherine Hermann, Ting Chen, and Simon Kornblith. The origins and prevalence of texture bias in convolutional neural networks. Advances in Neural Information Processing Systems, 33:19000-19015, 2020.","title":"The origins and prevalence of texture bias in convolutional neural networks","authors":"Katherine Hermann, Ting Chen, Simon Kornblith","year":"2020","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Jo and Bengio, 2017] Jason Jo and Yoshua Bengio. Mea- suring the tendency of cnns to learn surface statistical reg- ularities. arXiv preprint arXiv:1711.11561, 2017.","title":"Measuring the tendency of cnns to learn surface statistical regularities","authors":"Jo, Bengio, Jason Jo, Yoshua Bengio","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Krizhevsky et al., 2009] Alex Krizhevsky, Geoffrey Hinton, et al. Learning multiple layers of features from tiny im- ages. Handbook of Systemic Autoimmune Diseases, 1(4), 2009.","title":"Learning multiple layers of features from tiny images","authors":"Alex Krizhevsky, Geoffrey Hinton","year":"2009","venue":"Handbook of Systemic Autoimmune Diseases","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Schwarz et al., 2021] Katja Schwarz, Yiyi Liao, and An- dreas Geiger. On the frequency bias of generative models. Advances in Neural Information Processing Systems, 34, 2021. [","title":"On the frequency bias of generative models","authors":"Katja Schwarz, Yiyi Liao, Andreas Geiger","year":"2021","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Surasak et al., 2018] Thattapon Surasak, Ito Takahiro, Cheng-hsuan Cheng, Chi-en Wang, and Pao-you Sheng. Histogram of oriented gradients for human detection in video. In 2018 5th International conference on business and industrial research (ICBIR), pages 172-176. IEEE, 2018.","title":"Histogram of oriented gradients for human detection in video","authors":"Surasak","year":"2018","venue":"2018 5th International conference on business and industrial research (ICBIR)","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Wang et al., 2020] Haohan Wang, Xindi Wu, Zeyi Huang, and Eric P Xing. High-frequency component helps explain the generalization of convolutional neural networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8684-8694, 2020.","title":"High-frequency component helps explain the generalization of convolutional neural networks","authors":"Wang","year":"2020","venue":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Yun et al., 2019] Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo. Cutmix: Regularization strategy to train strong classifiers with localizable features. In Proceedings of the IEEE/CVF international conference on computer vision, pages 6023- 6032, 2019.","title":"Cutmix: Regularization strategy to train strong classifiers with localizable features","authors":"Yun","year":"2019","venue":"Proceedings of the IEEE/CVF international conference on computer vision","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Zhang et al., 2018] Hongyi Zhang, Moustapha Cisse, Yann N. Dauphin, and David Lopez-Paz. mixup: Beyond empirical risk minimization. In International Conference on Learning Representations, 2018.","title":"mixup: Beyond empirical risk minimization","authors":"Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, David Lopez-Paz","year":"2018","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}