CHAPTER 32 / 40
Modern Architectures & Decisions
3 lessons · Explore, experiment and understand.
DOJO PETYour little coding companion
DOJO / YOUR GUIDE
Let’s figure it out.
Ask me about this course or where to start.
Chapter lessons
3 lessons
- Step 134~15m3D · 113★Most cinematic
Attention and the Transformer
The core question. A sequence model must let each token gather information from other tokens. Recurrent networks (RNNs) do this by passing a…
Open lesson - Step 135~15m3D · 113★Most cinematic
Convolutional Networks: Locality, Sharing, and Output Shapes
Start with the problem that motivated convnets: images are large and structured. A modest RGB image is a vector in . A single fully-connecte…
Open lesson - Step 136~15m3D · 113★Most cinematic
RL Foundations: MDPs, Values, and Bellman
From prediction to control. Supervised learning answers "what is the label of this input?" — a one-shot, i.i.d. question. Reinforcement lear…
Open lesson