ScieNet: Deep Learning with Spike-assisted Contextual Information Extraction

TL;DR

ScieNet combines spiking neural networks with deep neural networks to enhance robustness against input perturbations like noise and rain, achieving high accuracy without prior training on perturbed data.

Contribution

The paper introduces a hybrid architecture integrating SNN-based unsupervised contextual extraction with DNNs, improving resilience to input perturbations in image classification.

Findings

Significant accuracy improvement on noisy images.

Maintains state-of-the-art accuracy on clean images.

Effective without prior training on perturbed data.

Abstract

Deep neural networks (DNNs) provide high image classification accuracy, but experience significant performance degradation when perturbation from various sources are present in the input. The lack of resilience to input perturbations makes DNN less reliable for systems interacting with physical world such as autonomous vehicles, robotics, to name a few, where imperfect input is the normal condition. We present a hybrid deep network architecture with spike-assisted contextual information extraction (ScieNet). ScieNet integrates unsupervised learning using spiking neural network (SNN) for unsupervised contextual informationextraction with a back-end DNN trained for classification. The integrated network demonstrates high resilience to input perturbations without relying on prior training on perturbed inputs. We demonstrate ScieNet with different back-end DNNs for image classification using CIFAR dataset considering stochastic (noise) and structured (rain) input perturbations. Experimental results demonstrate significant improvement in accuracy on noisy and rainy images without prior training, while maintaining state-of-the-art accuracy on clean images.