In this project, I aim to replicate the methodology and experiments presented in the paper "Group Equivariant Convolutional Networks" by Cohen and Welling and extend it to new applications, particularly focusing on evaluating its effectiveness in enhancing robustness against adversarial attacks.
