AI agents running research on single-GPU nanochat training automatically

Normalizing flows for neuro-symbolic AI

🚀🚀 「大模型」2小时完全从0训练64M的小参数GPT！🌏 Train a 64M-parameter GPT from scratch in just 2h!

Examples and tutorials to help developers build AI systems

This repo is meant to serve as a guide for Machine Learning/AI technical interviews.

About A collection of AWESOME things about information geometry Topics

Machine Learning Engineering Open Book

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Retrieval Augmented Generation (RAG) chatbot powered by Weaviate

Understanding Deep Learning - Simon J.D. Prince

General Purpose Risk Modeling and Prediction Toolkit for Policy and Social Good Problems

Master Federated Learning in 2 Hours—Run It on Your PC!

Information plane analysis of dropout networks

ICLR 2022, "FedBABU: Toward Enhanced Representation for Federated Image Classification"

A modern look at the relationship between sharpness and generalization [ICML 2023]

Experiments and implementations of the methods described in the paper "Relative Flatness and Generalization".

PyTorch implementation for the ICLR 2020 paper "Understanding the Limitations of Variational Mutual Information Estimators"

nanoGPT-like codebase for LLM training

Google's Operations Research tools:

Code for visualizing the loss landscape of neural nets

explore DNNs via Infomration

Implementation of Information Dropout

PyHessian is a Pytorch library for second-order based analysis and training of Neural Networks

ASDL: Automatic Second-order Differentiation Library for PyTorch

BackPACK - a backpropagation package built on top of PyTorch which efficiently computes quantities other than the gradient.

A tool to quantify and report the carbon footprint of machine learning computations and communication

Computing various measures and generalization bounds on convolutional and fully connected networks

Official code for "In Search of Robust Measures of Generalization" (NeurIPS 2020)

Code for the anonymous submission "Cockpit: A Practical Debugging Tool for Training Deep Neural Networks"

A curated list of awesome open source libraries to deploy, monitor, version and scale your machine learning