Night with a Nerd: Neuromorphic Artificial Intelligence - Why and where it might be useful

Neuromorphic Artificial Intelligence - Why and where it might be useful. Talk by Andrew Sornborger

Dr. Sornborger's talk will give an overview of neuromorphic computing, including the various analog and digital methodologies that investigators use to study how neural systems (neurons communicating with spikes over massively parallel networks) can be used to process information. He will then discuss a framework that he and collaborators have developed for controlling information flow and learning in neuromorphic computers. This framework makes it possible to easily determine where information propogates and when it gets to its target. He will then give examples of the types of AI algorithms that his group has produced and the value that their realization can have for AI "at the edge", in autonomous systems, like satellites, and in high-power-consuming installations such as data centers.

$10 admission; free for BSMA members (you can become a member on our Join Us page).

Wednesday, November 4th from 6-8pm at the Bradbury Science Museum (doors open at 6 and the talk starts at 6:30). 

About the Speaker

Andrew Sornborger received his PhD at Brown University in cosmology. After two postdocs, one at the University of Cambridge and a second at Fermilab, he switched fields to neuroscience and the analysis of neural data. After a postdoc and subsequent research assistant professorship at Mt Sinai School of Medicine in New York, NY, he moved to a tenure-track, then tenured joint position in mathematics and engineering at the University of Georgia. Following his wife to a new institution, he moved to the mathematics department at UC Davis. Then, finally, he came to LANL in 2017. Although hired for his neuromorphic computing expertise, his research was rapidly tilted toward quantum computing, a field he originally began working in while at Cambridge. Nonetheless, he still has a significant portfolio in neuromorphic computing, where he focuses on on-chip learning for artificial intelligence and signal processing applications.