# Feature Chirality in Deep Learning Models

arXiv: https://arxiv.org/abs/2305.03966 · HTML: https://mlsanity.com/paper/2305.03966/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 31 references that are in the map.
- First posted on arXiv 6 May 2023.

## Builds on (most cited first, 14 of 14)

- [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)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations](https://mlsanity.com/paper/1811.12359/) (2018)
- [One pixel attack for fooling deep neural networks](https://mlsanity.com/paper/1710.08864/) (2017)
- [Object Detection in 20 Years: A Survey](https://mlsanity.com/paper/1905.05055/) (2019)
- [Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients](https://mlsanity.com/paper/1711.09404/) (2017)
- [Interpretable Convolutional Neural Networks](https://mlsanity.com/paper/1710.00935/) (2017)
- [Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples](https://mlsanity.com/paper/1708.05493/) (2017)
- [Human-Aided Saliency Maps Improve Generalization of Deep Learning](https://mlsanity.com/paper/2105.03492/) (2021)
- [Visual Chirality](https://mlsanity.com/paper/2006.09512/) (2020)
- [Understanding deep features with computer-generated imagery](https://mlsanity.com/paper/1506.01151/) (2015)

## Abstract (arXiv)

As deep learning applications extensively increase by leaps and bounds, their interpretability has become increasingly prominent. As a universal property, chirality exists widely in nature, and applying it to the explanatory research of deep learning may be helpful to some extent. Inspired by a recent study that used CNN (convolutional neural network), which applied visual chirality, to distinguish whether an image is flipped or not. In this paper, we study feature chirality innovatively, which shows how the statistics of deep learning models' feature data are changed by training. We rethink the feature-level chirality property, propose the feature chirality, and give the measure. Our analysis of feature chirality on AlexNet, VGG, and ResNet reveals similar but surprising results, including the prevalence of feature chirality in these models, the initialization methods of the models do not affect feature chirality. Our work shows that feature chirality implies model evaluation, interpretability of the model, and model parameters optimization.