Andreas Moshovos is a Professor in the Department of Electrical and Computer Engineering at the University of Toronto's Faculty of Applied Science and Engineering. He holds a Bachelor's and Master's degree from the University of Crete, and a PhD from the University of Wisconsin-Madison. Previously, he taught at Northwestern University, École Polytechnique Fédérale de Lausanne, University of Athens, and Hellenic Open University. His research focuses on designing optimized computing hardware for performance, energy efficiency, and cost. Primary interests include: Deep Learning acceleration through value-based optimization (exploiting sparsity/precision variability) Hardware specialization for neural networks Efficient data management systems High-performance processor/memory architecture His publications demonstrate consistent focus on hardware acceleration techniques for deep learning, particularly leveraging value sparsity, dynamic precision adaptation, and ineffectual computation elimination to boost performance and energy efficiency in neural network processing. Significant Awards: ACM SIGARCH Maurice-Wilkes Award (2010) 2× IEEE MICRO Top Picks Awards (2006, 2010) 2× IBM Faculty Awards (2008, 2009) NSF CAREER Award (2000) MICRO Hall of Fame Award He leads the NSERC COHESA Network on Machine Learning Hardware Acceleration (19 researchers across 7 universities) and advises multiple PhD/Master's students in computer architecture and deep learning acceleration. His lab develops specialized hardware for computational imaging, gene sequencing, and neural network optimization.
