Abstract
Artificial intelligence allows us to solve increasingly complex problems, but algorithms require a lot of computational resources that consume a lot of power, especially during the learning phase to train neural networks with a large amount of data. Today, the need to process the data where they are generated requires low power architectures.This article presents a novel neural network based on oscillatory neural networks (ONN) which are inspired by the oscillations observed in the brain. ONNs present an alternative low power computing paradigm that enables online learning and inference on edge IoT devices.