معرفی
Setareh Behroozi is an Assistant Teaching Professor in the Department of Electrical and Computer Engineering at the University of Wisconsin–Madison. Her research focuses on energy-efficient computing, approximate hardware techniques, and hardware-software co-design for embedded systems and neural networks. She obtained her Ph.D. from the University of Wisconsin–Madison in 2022, an M.S. from Sharif University of Technology in 2015, and a B.S. from Iran University of Science and Technology in 2013.
Her publications emphasize scalable solutions for energy-quality tradeoffs in computing systems, including approximate dividers, neural network accelerators, and sensing hardware. These works aim to optimize energy efficiency without compromising critical system performance. Recent trends in her research include virtualizing nonlinear operations in neural networks and scheduling iterative hardware units for efficiency. She has contributed to both academic and applied aspects of low-power computing and embedded systems design.
Behroozi’s work bridges theoretical computer architecture with practical hardware implementations, addressing challenges in modern computing systems through innovative approximate computing techniques. She is actively involved in teaching and curriculum development within the Electrical and Computer Engineering department at UW-Madison.
