Yann Dubois

disentangling-vae. Experiments for understanding disentanglement in VAE latent representations

Neural-Process-Family. Code for the Neural Processes website and replication of 4 papers on NPs. Pytorch implementation.

Hash-Embeddings. PyTorch implementation of Hash Embeddings (NIPS 2017). Submission to the NIPS Implementation Challenge.

lossyless. Generic image compressor for machine learning. Pytorch code for our paper "Lossy compression for lossless prediction".

Invariant-Self-Supervised-Learning. Pytorch code for "Improving Self-Supervised Learning by Characterizing Idealized Representations"

SSL-Risk-Decomposition. Benchmark and analysis of 165 pretrained SSL models. Code for "Evaluating Self-Supervised Learning via Risk Decomposition".

Mini_Decodable_Information_Bottleneck. Minimum viable code for the Decodable Information Bottleneck paper. Pytorch Implementation.

YannDubs.github.io. Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes

miRNAseq-Analysis. Analyze single-cell miRNA sequencing data from HL-60 cells along a 7-day time-course of ATRA treatment.

Semi-Supervised-Neural-Processes. Jupyter Notebook

RAW-Embedings. Novel word embeddings based on a simple and intuitive rolling average. Still in dev mode.

simulationPopulationGenetics. A general modelisation of Wright Fisher in a context of total biological abstraction. And a specific modelisation of the evolution of the sickle cell anemia in Congo, Cameroon & Gaboon

lightning-bolts. Toolbox of models, callbacks, and datasets for AI/ML researchers.

npf. Experiments for comparing and understanding different neural processes architectures.

RubricEval. Jupyter Notebook

FourRussiansRNA. Faster algorithms for RNA-folding using the Four-Russians method

Infootmation. Visualise european soccer player's data

dino. PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

world-models. Reimplementation of World-Models (Ha and Schmidhuber 2018) in pytorch

Overlayed-Datasets. Code for generating datasets consisting of overlayed images.

TicTacToePyTorch. Tic tac toe dataset and PyTorch baseline.

optimizationCode. Some of the optimisation functions I've implemented: for learning purpose

machine-tasks. Datasets for compositional learning

vision. Datasets, Transforms and Models specific to Computer Vision

modellingBacterias. Useless project that tries to model the life of different bacterias in a petri dish: for learning purpose

phd-thesis-template. A LaTeX / XeLaTeX / LuaLaTeX PhD thesis template for Cambridge University Engineering Department (CUED)

QuoraQuestion. Siamese LSTM for Kaggle competition: Quora Question Pairs

scikit-optimize. Sequential model-based optimization with a `scipy.optimize` interface

awesome-demos. links and status of cool gradio demos

specialized-attention-mini. Mini Version of Specialized Attention