IEEE CAS WebinarAccelerating Streaming AI Inference using Sparsity and LocalityHost:    IEEE CAS Dallas chapter   (co-hosted by IEEE CAS SCV chapter)Date:Thursday, May 5, 2022Time:10:00 a.m. – 11:00 a.m. (PDT)Meeting: Online (see below for link to Zoom)

 

 Speaker:    Dr. Sam Fok, Founder and CEO, Femtosense

Abstract:

Femtosense enables streaming AI inference for up to 100x less energy and 10x less area than existing inference accelerators. AI algorithms demand an unprecedented level of compute. However, it is not the compute operations, but rather the underlying data motion operations that dominate energy costs. In this talk, we discuss how Femtosense employs sparsity and locality to mitigate energy costs. Sparsity is difficult to exploit on existing CPU and GPU-like architectures due to its irregular data access patterns. It is easy to lose any potential gains to the overhead necessary on existing architectures for sparse processing. Femtosense corrects this algorithm-architecture mismatch in its Sparse Processing Unit to enable large scale AI applications where previously considered impossible (e.g., the battery-powered edge).

Bio:

Sam is the CEO and co-founder of Femtosense where he is on a mission to bring human brain-level efficiency to artificial intelligence. He received BS degrees in biomedical engineering and electrical engineering from the Washington University in St. Louis and MS and PhD degrees in electrical engineering from Stanford University, where he was a member of the Brains in Silicon Laboratory. Sam and his co-founders started Femtosense out of the Brains in Silicon Lab with the purpose of commercializing the neuromorphic technology developed during their PhD research.   After a year of running into a brick wall of challenges associated with deploying mixed-signal subthreshold-analog-asynchronous-digital neuromorphic systems, they dispensed with the research-grade technology and pioneered the hybrid of fundamental neuromorphic principles and commercially viable all-digital architecture that Femtosense develops today.

Admission Fee:

All admissions free. Suggested donations:

Non-IEEE:  $5, Students (non-IEEE): $3, IEEE Members (not members of CASS or SSCS): $3

Date and time

Thursday, May 5th, 2022, 10:00 AM – 11:00 AM PST

Location

Online event

Register: Registration Link

Join Zoom Meeting:

https://us06web.zoom.us/j/83515600097?pwd=L281OTlEYlZMaGxIYUxIR2VNOHVEUT09

Meeting ID: 835 1560 0097

Passcode: 478866

———————;