A Low-Power RRAM Memory Block for Embedded, Multi-Level Weight and Bias Storage in Artificial Neural Networks.

Pattern recognition as a computing task is very well suited for machine learning algorithms utilizing artificial neural networks (ANNs). Computing systems using ANNs usually require some sort of data storage to store the weights and bias values for the processing elements of the individual neurons. This paper introduces a memory block using resistive memory cells (RRAM) to realize this weight and bias storage in an embedded and distributed way while also offering programming and multi-level ability.

Design of an Integrated Subretinal Implant using Cellular Neural Networks for Binary Image Generation in a 130 nm BiCMOS Process.

Blindness caused by the eye diseases Retinitis-Pigmentosa and Age-Related-Macular-Degeneration leads to a degeneration of the photoreceptor layer while postsynaptic cells mostly stay intact. In this Paper a new concept for retinal implants is proposed. Instead of converting the incident light to a gray-scale picture with corresponding continuous-value stimulation levels, we here suggest to produce a binary image picture that only highlight edges in order to stimulate the retina solely at points which belong to an edge.