Work

Analog CMOS-based resistive processing unit for deep neural network training

Year: 2017
Type: conference-paper
Abstract: Recently we have shown that an architecture based on resistive processing unit (RPU) devices has potential to achieve significant acceleration in deep neural network (DNN) training compared to today's software-based DNN implementations running on CPU/GPU. However, currently available device candidates based on non-volatile memory technologies do not satisfy all the requirements to realize the RPU concept. Here, we propose an analog CMOS-based RPU design (CMOS RPU) which can store and process data locally and can be operated in a massively parallel manner. We analyze various properties of the CMOS RPU to evaluate the functionality and feasibility for acceleration of DNN training.
Language: English
Open Access status: green
OpenAlex ID: W2675031852
FWCI: 11.25
Cites: 9
Cited by: 67
Related to: 10