Second Symposium on Algorithmic Information Theory and Machine Learning,
30 -31 July 2025, Imperial College London, London, UK
Following the success of the 1st edition held at the Alan Turing Institute, London, UK, on 4–5 July 2022 (see details), this symposium continues to explore the rich interface between Algorithmic Information Theory (AIT) and Machine Learning (ML).
Objective:
The symposium aims to bring together researchers working at the intersection of AIT and ML, with a particular focus on the application of concepts such as Kolmogorov Complexity and algorithmic probability to contemporary ML problems. Topics of interest include, but are not limited to:
Using AIT to explain a priori why certain ML methods succeed (or fail) on specific problems
Theoretical foundations and practical applications of Solomonoff induction
Connections between information theory, learning, and data compression
Developing ML algorithms for improved prediction and compression that approximate Kolmogorov Complexity
Probabilistic modeling through algorithmic probability frameworks
Kernel methods for AIT, and vice versa.
Organizing Committee: Boumediene Hamzi, Kamaludin Dingle, Marcus Hutter
Call for Participation:
If you are interested in giving a talk, please reach out to Boumediene Hamzi at bhamzi@turing.ac.uk.
Join Us:
Whether you plan to attend in person or online, or simply wish to stay informed about future events and updates on AIT & ML, please complete the interest form here.
Recordings:
Talk recordings will be made available after the event and posted here.
Food and Refreshments:
Please note that lunch will not be provided. Participants are encouraged to explore the diverse dining options available in and around Imperial College London. Light refreshments may be offered during the event; however, their availability is not guaranteed.
