Events Calendar

EE 598 Cyber-Physical Systems Seminar Series

Speaker: Ajay Jayant Joshi, Associate Professor, Boston University

Talk Title: Designing Energy-efficient and Secure Accelerators for Machine Learning Applications

Abstract: Today's mobile applications like activity tracking, photo/document sorting, fingerprint matching, search suggestions, etc. are increasingly data driven and commonly use Machine Learning (ML) algorithms. Executing these ML algorithms locally on the mobile system is sometimes preferable/necessary, but this local execution can be very energy intensive. At the same time, keeping these ML algorithm secure is becoming increasingly critical for application vendors as the use of the right ML algorithm can provide significant competitive (and in turn financial) advantage in the market. Hence, there is a need to execute these ML algorithms in an energy-efficient and secure manner. This talk focuses on the design of energy-efficient and secure hardware accelerators for ML-based applications. In the first half of my talk I'll present an adaptive classifier design that leverages the wide variability in data complexity to enable energy-efficient data classification operations. This adaptive classifier takes advantage of varying classification 'hardness' across data to dynamically allocate an appropriate classifier and improve energy efficiency. In the second half of my talk, I'll present a backside imaging approach that can be used to detect any insertion of Hardware Trojans during the fabrication phase. In particular, we engineer the fill cells in a standard cell library to be highly reflective at near-IR wavelengths so that they can be readily observed in an optical image taken through the backside of the chip. The pattern produced by their locations produces an easily measured watermark of the circuit layout. Any replacement, modification or re-arrangement of the fill cells to add a Hardware Trojan can therefore be detected through rapid post-fabrication backside imaging.

Biography: Ajay Joshi received his Ph.D. degree from the ECE Department at Georgia Tech in 2006. He then worked as a postdoctoral researcher in the EECS Department at MIT. In 2009, he joined the ECE department at Boston University, where he is currently an Associate Professor. His research interests span across various aspects of VLSI design including circuits and architectures for communication and computation. He received the NSF CAREER Award in 2012 and Boston University ECE Department's Award for Excellence in Teaching in 2014.

Host: Paul Bogdan

Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248

Audiences: Everyone Is Invited

Contact: Estela Lopez