The era of Generative AI is certainly upon us. In this talk, Professor Rosalyn Moran will present a theory of cortical function, and beyond, known as the Free Energy Principle which offers an alternative rationale and implementation for a Generative AI based on the brain. The Free Energy Principle, has been proposed as an ‘all-purpose model’ of the brain and human behavior that crucially closed the loop with action informing inference. As a formal and technical ‘first principles’ mathematical account of how brains work, it has garnered increasing attention from computer science to philosophy. The theory is based on the mathematical formulation of surprise minimization, to do so a brain can minimize its Free Energy (a computable bound on surprise), and drive, not only perception and cognition but crucially also actions. As a framework, the Free Energy Principle and its corollary ‘Active Inference’ thus represents a fundamental departure from current systems in Artificial Intelligence, as it calls for the implementation of a top-down system, rather than a bottom-up system (driven by masses of training data) that are currently the state-of-the-art frameworks in AI research. In this talk Professor Rosalyn Moran will demonstrate how we utilised the Free Energy Principle and Active Inference as an AI solution to simulated real-world problems.

If you are unable to attend in person, use this MS Teams link to attend the event virtually. 

Speaker

Rosalyn Moran a Professor of Computational Neuroscience at the IOPPN, King’s College London and the Deputy Director of King's Institute for Artificial Intelligence. Her work spans engineering and cognitive and computational neuroscience. In her lab she uses the Free Energy Principle as a principle to develop new methods in artificial intelligence and in modelling the brain’s normative and pathological function. She has previously held faculty positions at Virginia Tech and the University of Bristol.

This seminar is part of an ongoing seminar series from the Department of Engineering, King's College London.