[CTN] CTN Seminar - January 27 - Alex Ororbia

[Sue Ann Campbell]

Hello Everyone,

Our first CTN Seminar of the term will be held next week

Tuesday, January 27 at 3:30 p.m.  in DC 1304

Title:
Beyond Backpropagation of Errors: Predictive Coding, Biomimetic Intelligence, and NeuroAI

Speaker: Alex Ororbia, Rochester Institute of Technology https://www.cs.rit.edu/~ago/

Abstract:
Artificial intelligence (AI) is swiftly becoming one of the pivotal technologies of this century. Deep neural networks, trained by the backpropagation of errors algorithm, centrally drive many of AI’s impressive results. Nevertheless, the ubiquitous adoption of these kinds of models has highlighted some important limitations and practical shortcomings such as high energy inefficiency, lack of robustness, and biological implausibility. In addressing these issues, an interdisciplinary branch of AI research has emerged – biomimetic intelligence – which lies at the intersection of neuroscience, cognitive science, and machine learning, fostering the development of approaches motivated by neurocognitive theories. One such theory is known as predictive coding (PC), which has led to the development of a new class of computational neuronal models that have shown promising performance across various machine intelligence tasks.

This talk will foreground research in this direction, highlighting the many ways that PC might play a role in the future of machine learning, brain-inspired computing, and biomimetic intelligence in general. Next, we will explore promising efforts in casting PC’s neuronal dynamics in terms of spike trains; these lay the foundation for the development of real time, adaptive neuromorphic processing systems capable of addressing the prohibitive energy inefficiency inherent to modern-day neural architectures, including generative-pretrained transformers. To conclude, we finally consider the future challenges that lie ahead in making PC-based schemes more widely adopted in practical applications.

Bio:
Alex Ororbia is an Associate Professor in Computer Science and Cognitive Science, affiliate Professor in Psychology, and affiliate Professor in Mathematics at the Rochester Institute of Technology. He also serves as the director of the Neural Adaptive Computing Laboratory. His research group investigates questions related to neurobiological credit assignment and develops mathematical models for spiking neuronal dynamics and synaptic plasticity, predictive coding, and active inference with applications in neurorobotic control.

Hope to see you there,

Sue Ann


_______________________________________________


Dr. Sue Ann Campbell (she/her)

Professor and University Research Chair

Department of Applied Mathematics & Centre for Theoretical Neuroscience<https://uwaterloo.ca/centre-for-theoretical-neuroscience/>

Associate Dean, Research, Faculty of Mathematics

University of Waterloo

Waterloo ON N2L 3G1

https://www.math.uwaterloo.ca/~sacampbe/

Past-President, Canadian Applied and Industrial Mathematics Society<https://caims.ca/>


I acknowledge that I live and work on the traditional territory of the Chonnonton, Anishinaabeg and Haudenosaunee peoples. The University of Waterloo main campus is located on the Haldimand tract, the land granted to the Six Nations that includes six miles on each side of the Grand River. Our active work toward reconciliation takes place across our campuses through research, learning, teaching, and community building, and is co-ordinated within the Office of Indigenous Relations.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://artsservices.uwaterloo.ca/pipermail/ctn/attachments/20260121/739d7bcd/attachment.htm>