# Neural networks, attention and embeddings

> Build from artificial neurons and backpropagation to attention and embedding geometry.

Prerequisites: Vectors, matrix multiplication and functions.

Short self-study course; not an accredited qualification or a complete university syllabus.

## Lessons

- [Neural layers, activations and parameter counts](https://studybrix.com/en/courses/deep-learning/ai-neurons/index.md): university; 8 questions.
- [Backpropagation and the chain rule](https://studybrix.com/en/courses/deep-learning/ai-backprop/index.md): university; 8 questions.
- [Attention, softmax and transformer structure](https://studybrix.com/en/courses/deep-learning/ai-attention/index.md): advanced; 8 questions.
- [Embeddings, similarity and retrieval representations](https://studybrix.com/en/courses/deep-learning/ai-embeddings/index.md): university; 8 questions.