Home Banner

Our lab is focused on designing and creating efficient hardware systems to meet the growing computational needs of Machine Learning and Artificial Intelligence. We pursue a Physics-to-Systems approach where we combine algorithmic understanding with tailored hardware and architectures. We try to map unique features of emerging materials to the needs of algorithms and applications to build natural and energy-efficient computing systems. Our research interests lie at the intersection of physics, computer science and electrical engineering. 

We believe that such an interdisciplinary approach connecting architectures and algorithms to materials and devices is essential in the new era of electronics driven by domain-specific hardware. 

This is in contrast to Moore’s Law-driven era that has centered around a single device, the field-effect transistor. The new era of electronics calls for a new kind of scientist who needs to be deep in one field but also broad enough to be able to make connections to related disciplines across the computing stack. 


 Join OPUS Lab!

We are recruiting!

 Research Areas

We extend algorithms and architectures that match features of emerging hardware to cater to the needs of computation.
Designing efficient circuits with new functionalities often involves mapping materials directly to applications.
We seek to translate emerging materials and phenomena into physics-based circuit models to design benchmark circuits.

 Recent News

July 7, 2021

Andrea is investigating novel device opportunities for probabilistic computing. 

May 22, 2021

OPUS Lab receives NSF support to develop integrated nanodevices for probabilistic computing. 

April 28, 2021

for contributions to the theory and practice of using low barrier nanomagnets for probabilistic computing 

April 27, 2021

Sanaaya is working on alternative ways of executing probabilistic algorithms for p-computing. 

April 1, 2021

Waiting for Quantum Computing? Try Probabilistic Computing

(Illustration by Serge Bloch)

March 30, 2021

 "Naturally Probabilistic" Computing is highlighted by Purdue and UCSB ECE.  

(Illustration by Gwen Keraval)

March 25, 2021

Double-Free Layer MTJ concept appears in Physical Review Applied as Editor's Suggestion.

See the additional APS Physics coverage.


December 19, 2020

Navid Anjum Aadit joins our team as a PhD student. 

August 20, 2020

Kerem gives a virtual invited talk at the 31st Magnetic Recording Conference August 17-20, 2020 on "Probabilistic Computing using Stochastic Magnetic Tunnel Junctions".