Dissertation Title: Logic Synthesis for Low-Cost Stochastic Circuits and Approximate Circuits

Date: 2024/05/23 - 2024/05/23

Dissertation Title: Logic Synthesis for Low-Cost Stochastic Circuits and Approximate Circuits

Speaker: Xuan Wang, Ph.D. candidate at UM-SJTU Joint Institute

Time: May 23th from 9:30 a.m., 2024 (Beijing Time)

Location: Room 414A, Longbin Building

Abstract

With the advent of the post-Moore era, there are two issues in designing modern digital systems: improving the reliability and the energy efficiency of the circuits. Thus, two emerging computing paradigms, stochastic computing (SC) and approximate computing, are proposed to address these challenges. SC operates on stochastic bit streams, which shows strong fault tolerance to bit-flip errors and can realize complex arithmetic functions with simple circuits. Approximate computing designs circuits for error-tolerant applications. It introduces slight imprecision in computation to reduce the area, delay, and power consumption of the circuits. This dissertation studies logic synthesis methods for designing low-cost stochastic circuits and approximate circuits.

Biography

Xuan Wang received her B.S. degree in information engineering from Xidian University in 2018. She is currently a Ph.D. candidate at the UM-SJTU Joint Institute, supervised by Prof. Weikang Qian. Her research mainly focuses on logic synthesis for emerging computing paradigms, including approximate computing and stochastic computing.