Timezone isn't accessible, please provide the correct parameters
eventFeedUrl=http://realintelligence.com/customers/expos/00Do0000000aAt2/FMS_xmlcreator/a0J5c00001MW1eJ_specific-event-list.xml
trackCategory=Session
eventID=a0J5c00001MW1eJ
timezone=
duration=PTH
, NaNth
9:45-10:50 AM
AIAP-301-2: Neural Networks (Artificial Intelligence Applications Track)
Paper Title: Standard Logic Based Non-Volatile Neuromorphic Processor

Paper Abstract: Flash memory-based non-volatile neuromorphic processors have been proposed to efficiently perform dot-products for the growing intelligent AI applications running in edge devices. In this space is the single-poly embedded flash memory-based approach as a cost-effective solution for tiny neural network applications since it can be built in a generic logic process without any processing overhead. This presentation will propose an accurate analog programming method of the single-poly embedded flash memory cell that is demonstrated from the fabricated single-poly embedded flash memory based synaptic devices that have shown promising analog multiplication and accumulation results. This can be applicable to the cost-effective standard logic based non-volatile neuromorphic processor for energy constraint edge AI devices.

Paper Author: Sihwan Kim, Co Founder and Chief Product Officer, ANAFLASH

Author Bio: Seung-Hwan (Peter) Song is a CEO and Co-Founder of ANAFLASH Inc that develops a standard logic based non-volatile neuromorphic processor for smart edge AI devices. Dr. Song has more than 15 years of various experience from ANAFLASH, SK-HMS, Western Digital, Seagate, Qualcomm, Broadcom, and Samsung, and served as a principal investigator of the projects funded by NSF, NASA, and DoD Air Force. He won a best paper award from ICC 2016 and a low power design contest award from ISLPED 2012, and earned PhD from Univ. of Minnesota (ECE major and BME minor).