Navigation and service

PGI Colloquium: Prof. Dr. John Paul Strachan, Forschungszentrum Jülich (PGI-14), Germany; previously HP Inc., USA

Online Talk

Please note: You will receive the link to the online talk in the e-mail invitation, usually sent out a few days before the lecture takes place. It is also available on request from the contact person below.

begin
28 Jan 2022 11:00

Bringing neuromorphic principles and emerging device technologies into today’s computing systems

Bild

After approximately 100 years of engineering computers, humans have reached performance that rivals biological brains in many ways, while exceeding it in sheer number-crunching capacity. Yet, we took a very different path from biology, and there remains a huge advantage in energy-efficiency for biological information processing systems.

I will discuss our work to copy some of biology's tricks to build more efficient computers and tackle some of today’s hardest problems. This means re-visiting the design of computers from the bottom (devices) all the way up (algorithms), for which PGI is distinctly qualified. In one area, we are exploring computers augmented with “associative memories” for storing and retrieving complex patterns at low area and power consumption. Such an alternative to RAM speeds-up operations in genomics, security, and tree-based machine learning.

In another area, we use the stochastic analog operations in neural network dynamics to more quickly find the solutions to intractable Optimization problems, forecasting significant improvement over traditional and emerging compute technologies.

Contact

Dr. Felix Gunkel
Phone: +49 2461 61-5339
Fax: +49 2461 61-2550
email: f.gunkel@fz-juelich.de