Suyog Gupta Google, Mountain View, CA 94043 USA (suyoggupta@google.com). Dr. Gupta obtained his bachelor's and master's degrees in electrical engineering from IIT Bombay in 2009 and his Ph.D. degree in electrical engineering from Stanford University in 2013, where he worked on germanium-tin alloy for high performance CMOS and Si-compatible photonics applications. From 2013 to 2016, he was a Research Staff Member at the IBM Thomas J Watson Research Center working on novel hardware design for machine learning workloads and leading efforts on the use of approximate computing techniques and cross-layer system design for deep learning. The work reported in this paper was performed while he was with IBM. He is currently a Machine Learning Accelerator Architect at Google, working on on-device intelligence in consumer products.