What this resource explores
A structured introduction to deep learning, connecting mathematical foundations with neural network methods and research directions.
- Linear algebra, probability and optimization
- Neural networks and practical methodology
- Representation learning and research directions
Who it is for
Readers with mathematical and programming foundations who want a systematic view of deep learning.
Background & preparation
Linear algebra, probability and optimization recur throughout the text. Part I reviews mathematical and machine-learning foundations.
A way to approach it
Use Part I to identify gaps, Part II for practical neural-network methods, and Part III when you are ready to explore research directions.
Edition & context
This entry describes the 2016 text. It is not a current catalogue of every model family or software framework.