Adversarial examples to the new ConvNeXt architecture
To get adversarial examples to the ConvNeXt architecture, run the Colab: https://github.com/stanislavfort/adversaries_to_convnext/blob/main/adversaries_to_ConvNeXt.ipynb
To get more information, read the blog post: https://stanislavfort.github.io/blog/convnext_adversaries/
The A ConvNet for the 2020s paper from Facebook (Meta?) AI Research proposed a new architecture called ConvNeXt that is built out of standard convolutional blocks and seems to outperform the Vision Transformer (ViT). I wanted to know if it suffers from adversarial examples so I wrote a Colab that loads up a pretrained version of the ConvNeXt model, runs a quick loop of the Fast Gradient Sign Method, and demonstrates that it's easy to find adversaries to this new model. This is not surprising, but I thought it might be valuable to demonstrate it explicitly as well as to create a Colab that others can run and modify for their own experiments.
An example of what you can get from the Colab is this:
I am using the https://github.com/leondgarse/keras_cv_attention_models repository that made the whole process of loading the pretrained model really easy.