Learning, and the capstone

Working a behaviour out from data instead of writing it: fitting a model, judging it on data it has not seen, searching for a policy on the robot itself, what a reward function really specifies, the gap between the world you tuned in and the world you run in, and a capstone that uses the lot.

Module U12University7 lessons

1. U12.1 Learning a behaviour What learning buys over a hand-written rule, what it costs, and the smallest honest example.
2. U12.2 Fitting a model to data Least squares in plain Python, and a robot that drives by the model it fitted to itself.
3. U12.3 Generalisation Held-out data, overfitting, and choosing a model by the error it makes on points it never saw.
4. U12.4 Policy search on a robot Hill climbing on a real machine: a noisy score, a fair trial, and a budget measured in seconds.
5. U12.5 Reward is a specification The optimiser answers the question you asked, which is rarely the question you meant.
6. U12.6 The reality gap, and honest evaluation Why a policy tuned in one world fails in another, and how to report what it does without lying.
7. U12.7 Capstone: the whole robot Calibrate, estimate, localise, plan, follow, arrive, and report. One run, everything in it.