morganbdt Goto Github PK

alz_location_detect

crumb

Compositional Replay Using Memory Blocks (CRUMB) is a replay algorithm for online stream learning that stores and replays training examples at a feature level using a differentiable codebook to mitigate catastrophic forgetting. This repo also contains instructions for setting up the Toybox and iLab-2M-Light stream learning benchmarks.

gpm

Official Code Repository for "Gradient Projection Memory for Continual Learning"

lamp-python

learn-histopath

morganbdt.github.io

pytorch_data_loader_tutorial

remind

PyTorch implementation of the REMIND method from our ECCV-2020 paper "REMIND Your Neural Network to Prevent Catastrophic Forgetting"

robustness

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

scikit-learn

scikit-learn: machine learning in Python

toy_cl_demo

Example code that demonstrates how vanilla convolutional neural networks exhibit catastrophic forgetting in an incremental learning task

wavemixehr

An EHR topic model, based on MixEHR (Li et al. 2020), that incorporates ECG waveforms. By Payal Chandak, Daniel Shao, Joanna Shen, and Morgan Talbot (equal contribution)