The Linear Algebra of Deep Learning
summary
Explains how the core components of neural networks and deep learning are fundamentally operations from linear algebra. Covers matrices, vectors, and linear transformations as the basis behind layers, activations, and other network elements to show why seemingly complex or nonlinear behavior still relies on linear-algebraic building blocks.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:31Machine learning as linear algebra
- 3:00Key deep learning algorithms
- 5:53Singular value decomposition
- 12:01Example with satellite image
- 17:02Randomized SVD algorithm
- 20:45General matrix multiplication
- 26:24Derivatives in neural networks
- 28:53Automatic differentiation explained
- 31:42Application in satellite data
- 33:07Resources and closing
related talks
32mA Primer (or Refresher) On Linear Algebra for Data Science
43mThe Linear Algebra of Deep Learning
01h 00mAdventures in Linear Algebra++ and Unsupervised Learning
41mUnveiling the simple math that powers AI
51mDeep Learning - What is it and What It Can Do For You
34mThe physics of deep learning using tensor networks
54mIntroduction to Understanding Deep Learning
35mA Crash Course in Applied Linear Algebra