Timezone isn't accessible, please provide the correct parameters
eventFeedUrl=http://realintelligence.com/customers/expos/00Do0000000aAt2/FMS_xmlcreator/a0J1J00001H0ji2_specific-event-list.xml
trackCategory=Session
eventID=a0J1J00001H0ji2
timezone=
duration=PTH
, NaNth
2:10-5:00 PM
AIML-302-1: Using AI/ML for Flash Performance Scaling, Part 2 (AI/Machine Learning Track)
Paper Title: Computation and Machine Learning in Flash Memory: Opportunities and Challenges

Paper Abstract: Locally run machine learning algorithms,such as speech and image recognition are poised to transform human-machine interaction, but they are limited by power consumption. Analog neural networks promise dramatic advantages in power, throughput, and density over conventional digital implementations, enabling machine intelligence without the cost, latency, or privacy risks of cloud computing. Embedded flash memory is a strong contender for the weight storage and matrix-vector computation in analog neural networks, but presents several challenges, including temperature variation and high-voltage programming. We will describe the opportunities presented by analog neural networks, discuss fundamental limits and practical challenges, and compare existing and emerging memory technologies for neural network computation.

Paper Author: Jeremy Holleman, CTO, Syntiant

Author Bio: Jeremy Holleman is the the CTO of Syntiant and an Associate Professor of Electrical and Computer Engineering at the University of North Carolina, Charlotte. He received the B.S. degree in electrical engineering from the Georgia Institute of Technology, Atlanta, in 1997 and the M.S. and Ph.D. degrees in electrical engineering from the University of Washington, Seattle, in 2006 and 2009, respectively. From 2009 until 2016 he held the positions of Assistant and Associate Professor in the Department of Electrical Engineering and Computer Science at the University of Tennessee, Knoxville in 2009. In 2016 he joined the faculty of the University of North Carolina, Charlotte. He has previously worked for Data I/O and National Semiconductor. His research focuses on machine learning, mixed-mode computation, neuromorphic engineering, and ultra-low-power integrated circuits for biomedical devices and other wireless sensing applications.