Part XI: The AI Lab
From the chain rule to a small code model you train with reinforcement learning, reward with a grader you wrote, and then open up to see what it learned.
- AI00The AI Lab route: from the chain rule to a code model you trained and opened upDescribe all six modules and what you will build in each; say which new math arrives where and why it arrives there; know which tools each module installs, and which of them force WSL2; estimate your hours; know which code you…
- AI01Autograd from scratch: backpropagation is the chain rule25 exercisesCompute partial derivatives by freezing the other inputs; backpropagate by hand through any small graph, including values used more than once; explain why the backward pass runs in reverse topological order; build and test a…