Magnetic Tunnel Junction (MTJ) as Stochastic Neurons and Synapses: Stochastic Binary Neural Networks, Bayesian Inferencing, Optimization Problems

By Abhronil Sengupta1, Kaushik Roy2

1. Electrical Engineering and Computer Science, Penn State University, University Park, PA 2. Electrical and Computer Engineering, Purdue University, West Lafayette, IN

Published on

Abstract

Stochastic neuromorphic hardware, enabled by probabilistic switching of nanomagnets, has the potential of enabling a new generation of state-compressed, low-power brain-inspired computing platforms, which can be significantly more energy-efficient and scalable than their deterministic counterparts. Stochastic switching of devices also finds use in applications involving Deep Belief Networks and Bayesian Inference, as well as optimization problem solvers which utilize stochasticity to perform natural annealing. In this presentation, we provide a multi-disciplinary perspective across the stack of devices, circuits, and algorithms to illustrate how the stochastic switching dynamics of spintronic devices in the presence of thermal noise can provide a direct mapping to the units of such computing paradigms, namely Stochastic Binary Neural Networks, Bayesian Inferencing and Optimization Problems.

Bio

Abhronil Sengupta Abhronil Sengupta is an Assistant Professor in the School of Electrical Engineering and Computer Science at Penn State University. He is also affiliated with the Materials Research Institute (MRI). Abhronil received the PhD degree in Electrical and Computer Engineering from Purdue University in 2018 and the B.E. degree from Jadavpur University, India in 2013. He worked as a DAAD (German Academic Exchange Service) Fellow at the University of Hamburg, Germany in 2012, and as a graduate research intern at Circuit Research Labs, Intel Labs in 2016 and Facebook Reality Labs in 2017. He is pursuing an inter-disciplinary research agenda at the intersection of hardware and software across the stack of sensors, devices, circuits, systems and algorithms for enabling low-power event-driven cognitive intelligence.

Abhronil has been awarded the IEEE SiPS Best Paper Award (2018), Bilsland Dissertation Fellowship (2017), CSPIN Student Presenter Award (2015), Birck Fellowship (2013), the DAAD WISE Fellowship (2012), and his publications have featured as APL Editor?s Picks (2015) and top 5 popular articles in IEEE TCAS-I (2017). His work on spin-device based neuromorphic computing has been highlighted in media by MIT Technology Review, US Department of Defense, American Institute of Physics among others.

Kaushik Roy Kaushik Roy is the Edward G. Tiedemann, Jr., Distinguished Professor of Electrical and Computer Engineering at Purdue University. He received his PhD from University of Illinois at Urbana-Champaign in 1990 and joined the Semiconductor Process and Design Center of Texas Instruments, Dallas, where he worked for three years on FPGA architecture development and low-power circuit design. His current research focuses on cognitive algorithms, circuits and architecture for energy-efficient cognitive computing, computing models, and neuromorphic devices. Kaushik has supervised 75 PhD dissertations and his students are well placed in universities and industry. He is the co-author of two books on Low Power CMOS VLSI Design (John Wiley & McGraw Hill).

Kaushik received the National Science Foundation Career Development Award in 1995, IBM faculty partnership award, ATT/Lucent Foundation award, 2005 SRC Technical Excellence Award, SRC Inventors Award, Purdue College of Engineering Research Excellence Award, Humboldt Research Award in 2010, 2010 IEEE Circuits and Systems Society Technical Achievement Award (Charles Doeser Award), Distinguished Alumnus Award from Indian Institute of Technology (IIT), Kharagpur, Global foundries visiting chair at National University of Singapore, Fulbright-Nehru Distinguished Chair, DoD Vannevar Bush Faculty Fellow (2014-2019), Semiconductor Research Corporation Aristotle award in 2015.

Sponsored by

Cite this work

Researchers should cite this work as follows:

  • Abhronil Sengupta; Kaushik Roy (2018), "Magnetic Tunnel Junction (MTJ) as Stochastic Neurons and Synapses: Stochastic Binary Neural Networks, Bayesian Inferencing, Optimization Problems," http://nanohub.org/resources/29090.

    BibTex | EndNote

Time

Location

Purdue University, West Lafayette, IN

Tags