Neuromorphic computing using non-volatile memory (NVM) devices has the potential to enable extremely low power consumption, less than that of traditional von Neumann architecture. Low-power multiply-accumulate operations can be achieved using NVM crossbar arrays, where the conductance of an NVM device corresponds to a synaptic weight in a deep neural network (DNN) and spiking neural network (SNN). This talk covers the introduction of researches on neuromorphic computing using NVM devices.
Akiyo Nomura is a researcher in the Neuromorphic Device group at IBM Research - Tokyo. Her research interests include Neuromorphic computing, spintronics, and circuit design. For more detail, please visit her webpage: