We present a new image classification model that extends CNNs with biologically-inspired higher-order convolutions. Outperforms standard CNNs on benchmarks and shows unique representational properties, bridging neuroscience and deep learning.

We derive a power-law relationship between the number of examples per class and per-example vulnerability to membership inference and support it with extensive experiments.

Variational supervised contrastive learning maximizes a posterior-weighted ELBO, replacing pairwise comparisons with class-level interactions for SOTA performance on image classification tasks.

Interpretable Model with Global Interpretabiltiy and Novel Local Hierarchical Explanations, that can Provide Built-in Interpretable Coherent Set Prediction with Conformal Prediction