The Winter 2023 AIGSA reading group is about Foundations of Learning Theory. Join us for the first meeting! Lunch will be provided.

This week’s topic is **The Riddle of Induction**. Hume asks why we trust inference based on our past experiences. His argument later becomes formalized by Wolpert as part of the “No Free Lunch” theorems.

To prepare, **please spend up to 1 hour reading** at least Section 1 “Hume’s Problem” of the Stanford Encyclopedia of Philosophy entry on “The Problem of Induction”.

This term’s reading group topic is Foundations of Learning Theory. “Can my algorithm converge on the truth?” is an important question to ask when starting an AI problem, and a lot of theory has gone into proving an answer. We will explore one of these theoretical frameworks–Formal Learning Theory, which explores what can be learned from empirical data–as we work toward understanding the famous “No Free Lunch” theorems.

We will meet five times total this term, 1-2pm on alternating Wednesdays (Weeks 2, 4, 6, 8, and 10).

To prepare for each meeting, please spend ~1 hour on the week’s reading.

To serve as a counterexample to the “No Free Lunch” theorems, free lunch from Chipotle will be provided.