You wanna get into how ML works beneath the frameworks:
- "Understanding Deep Learning" by Simon J. D. Prince
- "The Hundred Pages Machine Learning Book" by Andriy Burkov
- "The Little Book of Deep Learning" by François Fleuret
You lack the maths skills to read the above:
- "Mathematics for Machine Learning" by Marc Peter Deisenroth, A. Aldo Faisal, Cheng Soon Ong
- "Linear Algebra Done Right" by Sheldon Axler
- "Linear Algebra" by 3Brown1Blue
You just wanna do ML:
You want to learn about LLMs
- Karpathy's youtube channel -- this has the most amount of signal, highest amount of alpha to learn about LLMs and building out GPT-2 level transformer models
Tips
- Do the exercises, always
- You're not learning math if you don't do the exercises
- You're not learning ML if you don't build models! Train them from scratch! Implement a multi-layer percepetron from scratch!
- Tinker after chapters/videos, tweak the examples or exercises, build simple models, build yet another MNIST classifier, but tinker!