When/Where: 23rd May, 2024, Noon-1PM. In-person event (Venue: EAG Laboratories – 810 Kifer Road, Sunnyvale)

(Use corner entrance: Kifer Road / San Lucar Court. Do not enter at main entrance on Kifer Road )

Registration: Link

If you face an issue with vtools registration send an email to hiuyung.wong at ieee.org to get the zoom link and indicate whether you are an IEEE member, IEEE EDS member, IEEE Student member

Contact: hiuyung.wong at ieee.org

Speaker: Dr. Glenn Ning

Abstract:

The von Neumann architecture’s intrinsic bottleneck in data transfer between processor and memory units hinders performance as data sets continue to grow. TetraMem’s memristive devices-based analog in-memory computing significantly boosts throughput and energy efficiency in deep learning.
Our approach utilizes pre-trained synaptic weights from cloud-based training, directly programming them into computing memristors/multi-level RRAMs made with nanometer thin-films for edge deployment and enabling post-tuning to accommodate specific scenarios. High-precision programmability ensures uniform performance across memristive networks by necessitating numerous distinguishable conductance levels in each device. This advancement benefits applications like neural network training and inference computing.
By achieving stable 8 bits and above multi-levels conductance in individual memristor devices (up to 11 bits/cell, as featured in Nature), we enable monolithically integrated semiconductor chips, featuring large crossbar arrays on complementary metal-oxide-semiconductor (CMOS) circuits in the commercial foundry, suitable for diverse AI applications. Our arbitrary precision computing based on analog computing work was recently published in Science…