An RRAM-based Hardware Implementation of a Radial Basis Function Neuron for Edge Classifiers
The work designs a novel RRAM-based cell, called Template piXeL (TXL), to create an efficient artificial neuron for edge devices. This cell uses analog memory to perform metric-based classification and online adaptation. It achieves low latency and high energy efficiency by simulating Radial Basis Function (RBF) neurons directly in hardware.