reading list for "Machine Learning II"

[gusl]

Nando de Freitas's new class has quite a remarkable selection of readings, both technical and "inspirational".

Larry Wasserman - All of Statistics

Luc Devroye, László Györfi, Gábor Lugosi - A Probabilistic Theory of Pattern Recognition

Cristian Gouriéroux, Alain Monfort - Simulation-Based Econometric Methods

Tor Norretranders - The User Illusion

Jeff Hawkins, Sandra Blakeslee - On Intelligence

Gerd Gigerenzer - Gut Feelings: The Intelligence of the Unconscious

and it sounds like we will also navigate much Geoff Hinton territory, including contrastive divergence learning. On the first lecture, Nando motivated many fascinating questions about the wisdom of crowds, learning by imagination (i.e. simulation-based learning), change blindness, cognitive coherence¹, learning by imitation, cultural transmission.

We will also have a different "scribe" every lecture, supposedly a common practice in Berkeley and Stanford. A scribe is a student who is responsible for taking notes for the entire class.

---

^{1 - by this I mean the phenomenon by which people make up explanations to justify their actions, post-hoc}