# What do neural networks learn in image classification? A frequency shortcut perspective

arXiv: https://arxiv.org/abs/2307.09829 · HTML: https://mlsanity.com/paper/2307.09829/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 44 references that are in the map.
- First posted on arXiv 19 Jul 2023.

## Builds on (most cited first, 27 of 27)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [On the Spectral Bias of Neural Networks](https://mlsanity.com/paper/1806.08734/) (2018)
- [A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI](https://mlsanity.com/paper/1907.07374/) (2019)
- [Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks](https://mlsanity.com/paper/1901.06523/) (2019)
- [Unmasking Clever Hans Predictors and Assessing What Machines Really Learn](https://mlsanity.com/paper/1902.10178/) (2019)
- [The Pitfalls of Simplicity Bias in Neural Networks](https://mlsanity.com/paper/2006.07710/) (2020)
- [Noise or Signal: The Role of Image Backgrounds in Object Recognition](https://mlsanity.com/paper/2006.09994/) (2020)
- [High Frequency Component Helps Explain the Generalization of Convolutional Neural Networks](https://mlsanity.com/paper/1905.13545/) (2019)
- [A Fourier Perspective on Model Robustness in Computer Vision](https://mlsanity.com/paper/1906.08988/) (2019)
- [Gradient Starvation: A Learning Proclivity in Neural Networks](https://mlsanity.com/paper/2011.09468/) (2020)
- [Unlearnable Examples: Making Personal Data Unexploitable](https://mlsanity.com/paper/2101.04898/) (2021)
- [Can contrastive learning avoid shortcut solutions?](https://mlsanity.com/paper/2106.11230/) (2021)
- [Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Survey](https://mlsanity.com/paper/1911.12116/) (2019)
- [Measuring the tendency of CNNs to Learn Surface Statistical Regularities](https://mlsanity.com/paper/1711.11561/) (2017)
- [Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU Models](https://mlsanity.com/paper/2103.06922/) (2021)
- [Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain](https://mlsanity.com/paper/2108.08487/) (2021)
- [Automatic Shortcut Removal for Self-Supervised Representation Learning](https://mlsanity.com/paper/2002.08822/) (2020)
- [An Impartial Take to the CNN vs Transformer Robustness Contest](https://mlsanity.com/paper/2207.11347/) (2022)
- [Deep frequency principle towards understanding why deeper learning is faster](https://mlsanity.com/paper/2007.14313/) (2020)
- [A Survey on the Robustness of Computer Vision Models against Common Corruptions](https://mlsanity.com/paper/2305.06024/) (2023)
- [Understanding Layer-wise Contributions in Deep Neural Networks through Spectral Analysis](https://mlsanity.com/paper/2111.03972/) (2021)

## Built on by (most cited first, 28 of 28)

- [Frequency-Spatial Entanglement Learning for Camouflaged Object Detection](https://mlsanity.com/paper/2409.01686/) (2024)
- [Fino1: On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance](https://mlsanity.com/paper/2502.08127/) (2025)
- [Can Biases in ImageNet Models Explain Generalization?](https://mlsanity.com/paper/2404.01509/) (2024)
- [Fourier-basis Functions to Bridge Augmentation Gap: Rethinking Frequency Augmentation in Image Classification](https://mlsanity.com/paper/2403.01944/) (2024)
- [Shortcut Learning in Medical Image Segmentation](https://mlsanity.com/paper/2403.06748/) (2024)
- [Out of Spuriousity: Improving Robustness to Spurious Correlations without Group Annotations](https://mlsanity.com/paper/2407.14974/) (2024)
- [United Domain Cognition Network for Salient Object Detection in Optical Remote Sensing Images](https://mlsanity.com/paper/2411.06703/) (2024)
- [Noise Matters: Optimizing Matching Noise for Diffusion Classifiers](https://mlsanity.com/paper/2508.11330/) (2025)
- [Towards Combating Frequency Simplicity-biased Learning for Domain Generalization](https://mlsanity.com/paper/2410.16146/) (2024)
- [Do ImageNet-trained models learn shortcuts? The impact of frequency shortcuts on generalization](https://mlsanity.com/paper/2503.03519/) (2025)
- [DFM-X: Augmentation by Leveraging Prior Knowledge of Shortcut Learning](https://mlsanity.com/paper/2308.06622/) (2023)
- [Language-Based Augmentation to Address Shortcut Learning in Object Goal Navigation](https://mlsanity.com/paper/2402.05090/) (2024)
- [Boosting Adversarial Transferability via High-Frequency Augmentation and Hierarchical-Gradient Fusion](https://mlsanity.com/paper/2505.21181/) (2025)
- [Hypernym Bias: Unraveling Deep Classifier Training Dynamics through the Lens of Class Hierarchy](https://mlsanity.com/paper/2502.12125/) (2025)
- [Visual Prompt-Agnostic Evolution](https://mlsanity.com/paper/2601.20232/) (2026)
- [Exploring connections of spectral analysis and transfer learning in medical imaging](https://mlsanity.com/paper/2407.11379/) (2024)
- [Lost in Context: The Influence of Context on Feature Attribution Methods for Object Recognition](https://mlsanity.com/paper/2411.02833/) (2024)
- [Target-Agnostic Calibration under Distribution Shift with Frequency-Aware Gradient Rectification](https://mlsanity.com/paper/2508.19830/) (2025)
- [On Measuring Localization of Shortcuts in Deep Networks](https://mlsanity.com/paper/2510.26560/) (2025)
- [Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification](https://mlsanity.com/paper/2601.10180/) (2026)
- [FiGKD: Fine-Grained Knowledge Distillation via High-Frequency Detail Transfer](https://mlsanity.com/paper/2505.11897/) (2025)
- [FOCUS: Frequency-Optimized Conditioning of DiffUSion Models for mitigating catastrophic forgetting during Test-Time Adaptation](https://mlsanity.com/paper/2508.14437/) (2025)
- [A Second-Order Perspective on Pruning at Initialization and Knowledge Transfer](https://mlsanity.com/paper/2509.24066/) (2025)
- [D-GAP: Improving Out-of-Domain Robustness via Dataset-Agnostic and Gradient-Guided Augmentation in Frequency and Pixel Spaces](https://mlsanity.com/paper/2511.11286/) (2025)
- [Plug, Play, and Fortify: A Low-Cost Module for Robust Multimodal Image Understanding Models](https://mlsanity.com/paper/2602.22644/) (2026)
- [Low-Frequency Shortcuts in Texture-Driven Visual Learning](https://mlsanity.com/paper/2606.03493/) (2026)
- [Invisible Shortcuts: Why Vision Encoders Know Your Camera](https://mlsanity.com/paper/2608.05424/) (2026)
- [Close Shortcut Wins Long: Seeking Diverse and Stable Generators for Data-Free Knowledge Distillation](https://mlsanity.com/paper/2608.22003/) (2026)

## Abstract (arXiv)

Frequency analysis is useful for understanding the mechanisms of representation learning in neural networks (NNs). Most research in this area focuses on the learning dynamics of NNs for regression tasks, while little for classification. This study empirically investigates the latter and expands the understanding of frequency shortcuts. First, we perform experiments on synthetic datasets, designed to have a bias in different frequency bands. Our results demonstrate that NNs tend to find simple solutions for classification, and what they learn first during training depends on the most distinctive frequency characteristics, which can be either low- or high-frequencies. Second, we confirm this phenomenon on natural images. We propose a metric to measure class-wise frequency characteristics and a method to identify frequency shortcuts. The results show that frequency shortcuts can be texture-based or shape-based, depending on what best simplifies the objective. Third, we validate the transferability of frequency shortcuts on out-of-distribution (OOD) test sets. Our results suggest that frequency shortcuts can be transferred across datasets and cannot be fully avoided by larger model capacity and data augmentation. We recommend that future research should focus on effective training schemes mitigating frequency shortcut learning.