Paper Title: Hardware Implementation of Differential Oscillatory Neural Networks Using VO2-Based Oscillators and Memristor-Bridge Circuits
Oscillatory Neural Networks (ONNs) can be potentially implemented with nano-scale oscillators coupled by resistors and capacitors, emulating traditional neural networks (NNs). However, many times negative interconnection weights are required, which is not obvious in ONNs.
Here, EU H2020 NeurONN CSIC IMSE Researchers present an elegant solution based on differential ONNs (DONNs) where each oscillator provides a pair of perfectly synchronized and opposite nodes, thus allowing negative weights by connecting positive resistances between opposite nodes of different nano-oscillators.
Authors: Jafar Shamsi, María José Avedillo, Bernabé Linares-Barrancoand Teresa Serrano-Gotarredona [Instituto de Microelectrónica de Sevilla (CSIC), Universidad of Sevilla, Seville, Spain]