JAX compilation of RDDL description files, and a differentiable planner in JAX.

A Deep Learning library that I'm writing for educational purposes.

A simple neural network for identifying handwritten digits using no high-level ml libraries (includes lots of math exposition)

NYCU Deep Learning Spring 2024

This repository contains my original code solutions and project reports, as well as the provided problem formulations for assignments 1, 2, 3 and 6 for the Natural Language Processing course, held in the Autumn semester 2022 at ETH Zürich. Each folder contains the files of the corresponding assignment.

Lightweight Python package for automatic differentiation

A modular C++17 framework for automatic differentiation

This project will give some highlight on the notion of F-adjoint which has been recently introduced in the following arxiv preprint: "Backpropagation and F-adjoint. arXiv preprint arXiv:2304.13820".

A basic neural network with backpropagation programmed from scratch in C++

neural network

This repository contains notes, slides, labs, assignments and projects for the Deep Learning Specialization by DeepLearning.AI and Coursera.

A simple, lightweight and powerful ML framework for Lua.

This library gives a modular design for better control of gradient passing between architecture components. Useful for architectures not using a traditional forward and backward pass.

In this project, we will construct deep learning models from scratch using NumPy , including linear regression, logistic regression, and neural networks. we also converge parameter estimation, back-propagation, and statistical inference.

scalar value gradient descent optimizer

Get Started with Deep Learning

From linear regression towards neural networks...

Understanding and Implementing Backprop Algorithm from scratch

TAGE: a Tensor-based Automatic Gradient Engine in python using NumPy for fast computation -> Ready for Neural Network and Gradient Descent Training