CISE Seminar: Yuval Cassuto, Israel Institute of Technology

Talk Title: Toward Reliable Distributed AI

A key question for AI performance and scalability is how to maintain its reliability when deployed across noisy channels and over faulty hardware. The challenge is that existing information-reliability measures are task independent, while AI quality is task dependent.

Our contributions are toward maintaining reliability of general inference tasks such as classification and regression. Such a task is performed jointly by an ensemble of distributed nodes, some (or all) of which may be noisy or faulty. The talk will present several approaches for reliable classification and regression in such settings. The strength of these approaches is that they build on a well-known and commonly used machine-learning primitive called “ensemble methods”, and thus they are applicable in great generality and flexibility.

Yuval Cassuto is a Professor at the Viterbi Department of Electrical and Computer Engineering, Technion – Israel Institute of Technology. His research interests lie at the intersection of the theoretical information sciences and the engineering of practical computing and storage systems. He has served on the technical program committees of leading conferences in both theory and systems. During 2010-2011 he has been a Scientist at EPFL, the Swiss Federal Institute of Technology in Lausanne. From 2008 to 2010 he was a Research Staff Member at Hitachi Global Storage Technologies, San Jose Research Center. In 2018-2019 he held a Visiting Professor position at Western Digital Research, and a Visiting Scholar position at UC Berkeley. He received the B.Sc degree in Electrical Engineering, summa cum laude, from the Technion in 2001, and the M.S. and Ph.D. degrees in Electrical Engineering from the California Institute of Technology, in 2004 and 2008, respectively. From 2000 to 2002, he was with Qualcomm, Israel R&D Center, where he worked on modeling, design and analysis in wireless communications. Dr. Cassuto has won the Best Student Paper Award in data storage from the IEEE Communications Society in 2010 as a student, and in 2019 as an adviser. He also won faculty awards from Qualcomm, Intel, and Western Digital. As an undergraduate student, he won the 2001 Texas Instruments DSP and Analog Challenge $100,000 prize.

Faculty Host: Ari Trachtenberg

Student Host: Chae Woo Lim