Foundation Model for Cell Painting (CellPainTR)
A novel contrastive learning framework to explicitly disentangle biological signals from nuisance variables (like batch effects) in high-content imaging data.
Mitigating Racial Bias in Face Age Progression
An investigation into fairness for generative models. We modified a StyleGAN-based age progression model to mitigate racial bias by promoting more equitable transformations across demographics.
A Deep Dive into Neural Style Transfer
An implementation and explanation of the seminal 2015 paper by Gatys et al., breaking down the concepts of content and style loss.
Interactive Histopathology Segmentation Playground
An educational toolkit with from-scratch PyTorch implementations of UNet, UNet++, and ResNet-UNet, complete with an interactive Streamlit app for exploring the PanNuke dataset.
The Geometry of Words: A Word2Vec Playground
A from-scratch PyTorch implementation that demystifies Word2Vec by exploring successes, corpus biases, and the fascinating geometry of language.
The Engine of Learning: A Guide to Backpropagation
A from-scratch guide that demystifies the core algorithm of deep learning. Build the backpropagation algorithm from a single neuron to a full network, with interactive exercises.