IBM NorthPole: An Architecture for Neural Network Inference with a 12nm Chip
- 2024
- ISSCC 2024
Megumi Ito is a research staff member at IBM Research - Tokyo since 2008. Her research interests include accelerators, performance optimizations, and performance analysis. She received her M.S. degree in computer science at University of Tsukuba. She has been working on performance analysis optimization with an FPGA accelerator for genome analysis applications using POWER8 CAPI feature, an FPGA accelrator for sorting, and a simulator for an analog neuromorphic chip. Her recent research focuses on developing new AI chip, designing circuit logic and verifying the architecture.