Code for the TMLR 2023 paper "Training DNNs Resilient to Adversarial and Random Bit-Flips by Learning Quantization Ranges".

Avalanche: an End-to-End Library for Continual Learning based on PyTorch.

Code associated with the paper "Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists"

Reduce gender bias in machine learning models.

Reimplementation of AAAI21: Parameter Corruption

The net:cal calibration framework is a Python 3 library for measuring and mitigating miscalibration of uncertainty estimates, e.g., by a neural network.

A simple way to calibrate your neural network.

Wrapper for a PyTorch classifier which allows it to output prediction sets. The sets are theoretically guaranteed to contain the true class with high probability (via conformal prediction).

Run a conference from your backyard.

Generative Flow Networks

IBM Analog Hardware Acceleration Kit

Use lmdb to speed up imagenet dataset

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

Code for the paper "Adversarially Regularized Autoencoders (ICML 2018)" by Zhao, Kim, Zhang, Rush and LeCun

Multilingual Sentence & Image Embeddings with BERT

BERT score for text generation

DNN quantization with outlier channel splitting

Prescribed Generative Adversarial Networks

SGD with compressed gradients and error-feedback: https://arxiv.org/abs/1901.09847

Sparse learning library and sparse momentum resources.

A PyTorch implementation of EfficientNet

Sparsified SGD with Memory: https://arxiv.org/abs/1809.07599

Code for the signSGD paper

meProp: Sparsified Back Propagation for Accelerated Deep Learning (ICML 2017)