Learning and Belief Revision: an Overview

ABSTRACT

Successful learning can be understood as convergence to true beliefs. Can belief revision policies (as studied in Knowledge Representation) generate sensible learning methods? The same can be asked about multi-agent belief revision, where a group of agents revise their collective conjectures by a combination of belief revision and belief merge. Finally, how could computational models of learning, such as neural networks, fit in that picture? In my talk I will address those questions using a mix of methods of (modal) logic and formal learning theory.

SHORT BIO:

Nina Gierasimczuk is an associate professor at the Department of Applied Mathematics and Computer Science at Technical University of Denmark (DTU Compute). Before, she was employed at the Institute for Logic, Language and Computation at the University of Amsterdam and the Bernoulli Institute of Mathematics, Computer Science and Artificial Intelligence at the University of Groningen. At present, she serves on the boards of the European Association for Logic, Language and Information and the Scandinavian Logic Society. She is also a member of the steering committees of the DaLi Workshop and the NMR Workshop, and an associate editor at Journal and Logic, Language and Information, Annals of Mathematics and Artificial Intelligence, Fundamenta Informaticae, and Nordic Machine Intelligence.

Her main research interest lies in the logical aspects of learning in both single- and multi-agent context, and involves knowledge representation, dynamic modal logic, computability theory, belief revision, and multi-agent systems. She is also studying the role of logic and logical modeling in cognitive science.

Click here to register via Zoom – it’s free!

Women in Logic https://womeninlogic.org/