Arnon Amir of IBM Research Almaden

NorthPole is a brain-inspired, silicon-optimized chip architecture suitable for neural inference. NorthPole intertwines compute and processing on a single chip and has no off-chip memory, no centralized memory. The novel architecture, first presented in the November 2023 issue of Science, mitigates the von Neumann bottleneck which impairs new performance and power advances of today's processors. It leads to high throughput, low latency, and excellent power efficiency. The talk will include an overview of the architecture and its key design principles. Then I will present measured energy efficiency, latency and throughput results and a comparison with GPU, TPU and other processors on three applications: The ResNet50 benchmark image classification network; the Yolo-v4 object detection network; and the Granite 3b Large Language Model (LLM). These results demonstrate the leapfrog advantage of novel brain-inspired architectures.

Biography:

Arnon Amir is a Senior Research Scientist in the Brain Inspired Computing group at the IBM Almaden Research Center, where he currently works on the NorthPole chip. He earned his B.Sc. Magna Cum Laude in electrical and computer engineering from the Ben-Gurion University, Beer Sheva, and his M.Sc. and D.Sc. in computer science from the Technion—Israel Institute of Technology, Haifa. Over the last 25 years since joining IBM Research he worked on a wide range of topics. Prior to NorthPole he worked on the TrueNorth neuromorphic chip where he developed the Corelet Programming Language and an event-based gesture recognition system using a Dynamic Vision Sensor (DVS) camera. In 2008 he started and led the DuraBytes project and was the visionary and co-inventor of the Linear Tape File system (LTFS) that received an Engineering Emmy Award from the Academy of Television Arts and Sciences. Dr. Amir has published more than 100 peer-reviewed papers in journals and conferences and more than 50 US patents. He was awarded the 1997/98 post-doctoral Rothschild fellowship and several IBM awards for research and innovation. Dr. Amir is a senior member of the IEEE.

David Yonce of Cogwear

Cogwear is a brain technology company that improves health, peace, and performance through comfortable, wearable technology for better brain insights. The Philadelphia-based start-up has developed a platform solution with AI-powered algorithms, giving people on-demand access to their own clinical-grade cognitive and emotional data in real time in virtually any setting. Pioneered in collaboration with neuroscience experts from the University of Pennsylvania and supported by the National Science Foundation and the National Institutes of Health, Cogwear’s proprietary dry electroencephalography (EEG) sensors comfortably and accurately measure, interpret, and report electrical brain activity through a secure app and shareable dashboard. Cogwear has a positive impact on society by supporting brain health, boosting performance, and driving better decisions.

In this talk, we will cover:

• The case for why having better brain measures is important
• The science behind what brain insights tell us and case studies on how to apply these insights
• Information on the Cogwear Platform
• Current Cogwear activities and partnerships

Biography:

David Yonce is the CEO of Cogwear, a brain health company that improves health, peace and performance through its comfortable, wearable technology for better brain insights.

David is a medical device and technology executive who has positioned companies for double-digit growth and acquisition, helped deliver over 50 new products to market, and has received 54 patents.

With nearly 25 years in the medical device space, including 20 of those in a management or executive capacity, he has held diverse roles across R&D, Operations, Marketing, Strategy, and Sales at companies such as Boston Scientific, AMS, and DSM Biomedical. He also serves on the advisory boards of several companies and counsels industry advocacy groups and multiple public seed fund initiatives.

David earned a B.S. in Electrical Engineering from Cornell University, holds an M.S. in Electrical Engineering from the University of Maryland, and graduated with honors with an M.B.A. from the University of Chicago Booth School of Business.