Loading Events

« All Events

  • This event has passed.

CESG SEMINAR: Towards General-Purpose Unary (Stochastic) Computing Architectures

April 16 @ 9:20 am - 10:10 am

Free

Dr. Joshua San Miguel
Assistant Professor, Department of Electrical and Computer Engineering
University of Wisconsin-Madison

 

TitleTowards General-Purpose Unary (Stochastic) Computing Architectures


Talking Points
:

– Primer of unary and stochastic computing

–  Unary computing architecture for GEMM applications

– Metrics for evaluating the stability and correlation of unary bitstreams

– UnarySim open-sourced simulator for unary architectures


Abstract
:

There is a growing need for ultra-low-power processor systems that are not only capable of fitting in tiny form factors (e.g., wearables, implants), but also general-purpose enough to run a diverse set of applications under stringent power budgets. To build such systems, we make the case for unary computing, with particular focus on stochastic computing. This paradigm uses unary logic and datatypes to shrink circuits by orders of magnitude; for example, a traditional multiplier circuit reduces to a single AND gate in unary. Though promising, the barrier-to-entry for unary computing is currently too high. Existing unary designs tend to be overspecialized for specific use cases, and there is a lack of standard toolchains and metrics for navigating the ad hoc unary design space. This talk will offer a primer on unary computing and present our recent work towards making unary computing more general-purpose and broadly accessible. Namely, we developed uGEMM, a unary architecture for general matrix multiplication applications, as well as new metrics and simulation tools for better understanding the complex properties of unary and stochastic systems (e.g., stability, correlation). This talk will also give a glimpse of our ongoing work towards a general-purpose unary CPU.


Biography
:

Dr. Joshua San Miguel is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Wisconsin-Madison. His research spans broadly across topics in computer architecture and systems, with recent focus on approximate computing, stochastic computing, interconnection networks and intermittent computing. He is a recipient of the NSF CAREER Award in 2021. His work has garnered several paper awards and three IEEE Micro Top Picks citations in 2016, 2017 (honorable mention) and 2021 for his contributions to branch prediction, approximate computing and unary computing, respectively. He received his PhD from the University of Toronto in 2017.

 

 

 

 

 

 

Zoom Meeting ID: 963 4348 1647
https://tamu.zoom.us/j/96343481647 

Details

Date:
April 16
Time:
9:20 am - 10:10 am
Cost:
Free
Website:
https://tamu.zoom.us/j/96343481647