Universal Algorithmic Intelligence

Aram Ebtekar is an algorithmic information theorist and former MATS scholar who has used mathematics to make progress on hard philosophical problems from the arrow of time to AI safety to physics. He will give a talk on algorithmic thermodynamics at the regular research meeting this Monday Oct. 6th at 3 pm ET (note updated time!).

Title: Foundations of algorithmic thermodynamics
Abstract: Many foundational physics questions hinge on the definition of entropy. The Boltzmann and Gibbs-Shannon entropies depend on a choice of macrovariables or ensemble, respectively. Bennett, Zurek, and Gacs instead propose the algorithmic entropy as a fundamental measure of a state’s information content. Modernizing their ideas with tools from stochastic thermodynamics leads to a computability-based theory with very general fluctuation theorems. Whereas the Gibbs-Shannon entropy gives the mean capacity to do work from a state ensemble that is known a priori, the algorithmic entropy determines a state’s capacity for work in a more objective sense.

Zoom link: https://uwaterloo.zoom.us/j/7921763961?pwd=TDatET6CBu47o4TxyNn9ccL2Ia8HN4.1

(The Q&A with David Quarel on the same day is part of the 12 pm ET reading group, 3 hours earlier, NOT the research meeting)