University of Pennsylvania CIS 7000
Professor Roth · University of Pennsylvania · 1 lecture with notes
Students in this class: ask your lecturer for the class code, and these lectures will already be in your library when you sign up.
Mathematical Foundations of AI Alignment, Lecture 11.2 (Tractable Agreement From Calibration)
Conversation calibration makes persistent human–AI disagreement costly in prediction error, yielding fast approximate agreement without Bayesian agents or a known prior.