Skip to content

Neural Networks: Zero to Hero

Description

  • Instructor: Andrej Karpathy
  • Prerequisites: Basic Python programming and some familiarity with deep learning concepts
  • Programming Language: Python
  • Difficulty: 🌟🌟🌟🌟
  • Class Hours: Approximately 19 hours

This hands-on deep learning course, taught by Andrej Karpathy, provides a detailed and intuitive introduction to neural networks and their underlying principles. The course starts with foundational concepts such as backpropagation and micrograd before progressing to building language models, WaveNets, and GPT from scratch. The emphasis is on practical implementation, with step-by-step coding explanations to help students understand and build complex models from the ground up.

Instructor Information

Andrej Karpathy is a renowned AI researcher and educator with extensive experience in deep learning and neural networks. He was the Senior Director of AI at Tesla, leading the computer vision team for Tesla Autopilot from 2017 to 2022. Prior to that, he was a research scientist and founding member at OpenAI (2015-2017). In 2023, he returned to OpenAI, contributing to improvements in GPT-4 for ChatGPT. In 2024, he founded Eureka Labs, an AI+Education company.

Karpathy holds a PhD from Stanford University, where he worked on convolutional and recurrent neural networks with Fei-Fei Li. He has collaborated with leading AI researchers, including Daphne Koller, Andrew Ng, Sebastian Thrun, and Vladlen Koltun. He also taught the first deep learning course at Stanford, CS 231n: Convolutional Neural Networks for Visual Recognition, which became one of the largest classes at the university.

Course Resources

  • Lecture Videos: YouTube Playlist
  • Assignments: Self-guided projects and code implementation exercises available throughout the lectures

For more information, watch the full playlist on YouTube.