EigenFold is a diffusion generative model for protein structure prediction (i.e., known sequence -> distribution of structures). It is based on harmonic diffusion, which incorporates bond constraints in the diffusion modeling framework and results in a cascading-resolution generative process.

TopoOpt is a novel DNN training system that co-optimizes the distributed training process across computation, communication, and network topology.

These are notebooks for reproducing the paper Computer Vision with a Single (Robust) Classifier

A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.

Improving Deep Network Debuggability via Sparse Decision Layers 

ADE20K is composed of more than 27K images from the SUN and Places databases. Images are fully annotated with objects, spanning over 3K object categories. Many of the images also contain object parts, and parts of parts. We also provide the original annotated polygons, as well as object instances for amodal segmentation. Images are also anonymized, blurring faces and license plates.