The Lie Derivative for Measuring Learned Equivariance

October 06, 2022 ยท Entered Twilight ยท ๐Ÿ› International Conference on Learning Representations

๐Ÿ’ค TWILIGHT: Eternal Rest
Repo abandoned since publication

Repo contents: .gitignore, .gitmodules, LICENSE, README.md, assets, exps_e2e.py, exps_layerwise.py, lee, pytorch-image-models, requirements.txt, stylegan3, sweep_configs

Authors Nate Gruver, Marc Finzi, Micah Goldblum, Andrew Gordon Wilson arXiv ID 2210.02984 Category cs.LG: Machine Learning Cross-listed cs.AI, cs.CV, stat.ML Citations 57 Venue International Conference on Learning Representations Repository https://github.com/ngruver/lie-deriv โญ 40 Last Checked 1 month ago
Abstract
Equivariance guarantees that a model's predictions capture key symmetries in data. When an image is translated or rotated, an equivariant model's representation of that image will translate or rotate accordingly. The success of convolutional neural networks has historically been tied to translation equivariance directly encoded in their architecture. The rising success of vision transformers, which have no explicit architectural bias towards equivariance, challenges this narrative and suggests that augmentations and training data might also play a significant role in their performance. In order to better understand the role of equivariance in recent vision models, we introduce the Lie derivative, a method for measuring equivariance with strong mathematical foundations and minimal hyperparameters. Using the Lie derivative, we study the equivariance properties of hundreds of pretrained models, spanning CNNs, transformers, and Mixer architectures. The scale of our analysis allows us to separate the impact of architecture from other factors like model size or training method. Surprisingly, we find that many violations of equivariance can be linked to spatial aliasing in ubiquitous network layers, such as pointwise non-linearities, and that as models get larger and more accurate they tend to display more equivariance, regardless of architecture. For example, transformers can be more equivariant than convolutional neural networks after training.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning