archival

We fix unwanted biases in standard deep anomaly detection with a new architecture. Extended from IEEE MLSP (2020) version.

A geometric scattering based network with learnable scale parameters

We propose a method called integrated diffusion for combining multimodal data, gathered via different sensors on the same system, to …

A fast diffusion-based earth mover’s distance

Optimal transport has recently been reintroduced to the machine learning community thanks in part to novel efficient optimization …

Single-cell RNA-sequencing (scRNA-seq) is a powerful tool to quantify transcriptional states in thousands to millions of cells. It is …

Biomolecular graph analysis has recently gained much attention in the emerging field of geometric deep learning. While numerous …

We fix unwanted biases in standard deep anomaly detection with a new architecture.

A Combination of Continuous Normalizing Flows and Dynamic Optimal Transport to model the dynamics of cells.

While neural networks are powerful approximators used to classify or embed data into lower dimensional spaces, they are often regarded …