Hiroshi Ando

Papers

Total Citations

H-Index

About

Hiroshi Ando is a pioneering researcher at the intersection of neuromorphic engineering and analog-digital VLSI design for real-time image recognition. His most influential work, the 2003 paper "A Convolutional Neural Network VLSI for Image Recognition Using Merged/Mixed Analog-Digital Architecture," introduced a groundbreaking hardware implementation that merged analog and digital processing to efficiently execute convolutional neural networks. This early contribution laid critical groundwork for energy-efficient, on-chip AI accelerators, anticipating the modern push toward edge computing. With 29 citations, the paper remains a touchstone for researchers exploring mixed-signal neural network hardware. Ando’s key contributions lie in demonstrating how analog circuits can perform core CNN operations—such as convolution and pooling—with dramatically lower power consumption than purely digital designs, while maintaining sufficient accuracy for real-time tasks. His work has influenced subsequent efforts in neuromorphic chips and low-power vision systems. For students and researchers, Ando’s research exemplifies the power of cross-domain thinking, blending circuit design, machine learning, and computer architecture to create hardware that is both biologically inspired and practically deployable.