Graphical Models for Neuromorphic Algorithm Design

Graphical Models for Neuromorphic Algorithm Design

Presenter

  • Kathleen Hamilton, Quantum Computing Institute
December 14, 2017 - 11:00am to 12:00pm

Abstract 

In this seminar, I will present novel neuromorphic algorithms, derived using graphical models and spin-glass physics concepts. This approach to algorithmic design explores the feasibility of using neuromorphic architecture to solve general optimization problems. In particular I will demonstrate how spiking neural systems can be constructed and used to implement a spike-based label propagation method for community detection in undirected, unweighted graphs. This approach can identify uniform and nonuniform communities with accuracies near 100% for random graphs with more than 100 vertices and known ground truths. 

Additional Information 

About the Speaker:

 

Kathleen Hamilton is a postdoctoral researcher in the Quantum Computing Institute working on algorithmic design for next-generation computing platforms. She has worked on developing methods for near-term quantum annealers and near-term neuromorphic processors. She studied strongly correlated electron systems at UC-Riverside, where she obtained her PhD in physics.

Sponsoring Organization 

Quantum Computing Institute

Location

  • Research Office Building
  • Building: 5700
  • Room: L-202

Contact Information

Share