AI Team
Members of the AI team can work on a few different projects, or all of them!
Students can begin by studying the book “Neural Networks From Scratch” by Harrison Kinsley and Daniel Kukieła. The team will work through the book’s content, implementing the concepts in Python, and will also explore additional resources to deepen their understanding of neural networks.
- Neural Networks From Scratch - The book by Harrison Kinsley and Daniel Kukieła.
- Neural Networks From Scratch GitHub Repository - The official GitHub repository for the book, containing code examples and exercises.
- Neural Networks From Scratch animations - Animations for core concepts in the book.
- Repository of AI books. Some optional books that may be interesting.
After getting a basic understanding of how neural networks work, we will implement a few. We will do the MNIST digit recoginition tutorial, both using PyTorch and from numpy, in the style of NNFS. Then students can move on to training or using other neural networks, such as convolutional neural networks (CNNs) for image classification tasks, or recurrent neural networks (RNNs) for sequence prediction tasks.
Students can also work on projects that involve programming AI agents by making web requests to APIs, such as OpenAI’s GPT-3 or other AI services. This can include building chatbots, text generation tools, or other applications that leverage AI capabilities. For these tasks we will use Github’s new Models API, which allows us to use AI models without needing to run them locally.