Name: Christopher Beckham, PhD
Type: User
Company: @mila-iqia
Bio: Research interests: few-shot learning, generative modelling, principled evaluation. ML scientist @AlpacaML @Waikato alumnus. @mila-iqia
Twitter: chris_j_beckham
Location: Montréal, Québec
Blog: http://www.beckham.nz
Christopher Beckham, PhD's Projects
acgan
Adversarially learned inference in Lasagne
Public repo for Augmented Multiscale Deep InfoMax representation learning
Official adversarial mixup resynthesis repository
amr_demo
Code for the paper "Adversarially Regularized Autoencoders (ICML 2018)" by Zhao, Kim, Zhang, Rush and LeCun
Convert a .json file specifying hyperparameters to individual shell scripts to be executed
dna barcoding
yeah nah
Pretty and useful exceptions in Python, automatically. Modified to pretty print Torch tensors.
Implemented algorithms from UC bioinformatics algorithms course
[MSG-GAN] Any body can GAN! Highly stable and robust architecture. Requires little to no hyperparameter tuning.
Official repository for "Overcoming challenges in leveraging GANs for few-shot data augmentation", accepted to CoLLAs 2022.
dropdown picker/launcher for mac os
R package that plots all possible histograms, scatterplots, and boxplots determined by the type of variables present in the specified dataset. Has two modes - can output graphs externally to png files, or simply display to the graphics device (the latter of which could be useful if producing R Markdown scripts).
Examples for my christorch repo
Cleverbot API from maxvitek, modified to work on SL4A
A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning
Official repository for CLEVR: Mental Rotation Tests
A Theano implementation of a CNN DSEBM (deep structured energy-based model) described in https://arxiv.org/pdf/1605.07717v2.pdf
All my presentations/reports + final project for the COMP767 reinforcement learning class at McGill.
Official code for paper: Conservative objective models are a special kind of contrastive divergence-based energy model
Relevant papers in Continual Learning
Implementation of the Conway-Maxwell-Binomial distribution (for unimodal/ordinal classification problems) in PyTorch.