Floating gate architecture for deep neural network application

摘要

A resistive processing unit (RPU) circuit for use in a neural network application includes at least one floating gate storage device, the floating gate storage device including a floating gate, a control gate and an inject/erase gate. The RPU circuit further includes a feedback circuit connected with the floating gate storage device. The feedback circuit is configured to maintain a substantially constant floating gate potential of the floating gate storage device during an update mode of operation of the RPU circuit, and is disabled during a readout mode of operation of the RPU circuit.