Improving Analog Neural Network Robustness: A Noise-Agnostic Approach with Explainable Regularizations

TL;DR

This paper introduces a hardware-agnostic, explainable regularization framework that significantly improves the robustness of deep analog neural networks against hardware noise, addressing a key challenge in analog signal processing devices.

Contribution

It presents a novel, explainable regularization method that enhances noise robustness in deep neural networks without relying on specific hardware assumptions.

Findings

Enhanced noise resilience in deep neural networks

Demystified mechanisms underlying noise robustness

Applicable across various analog hardware platforms

Abstract

This work tackles the critical challenge of mitigating "hardware noise" in deep analog neural networks, a major obstacle in advancing analog signal processing devices. We propose a comprehensive, hardware-agnostic solution to address both correlated and uncorrelated noise affecting the activation layers of deep neural models. The novelty of our approach lies in its ability to demystify the "black box" nature of noise-resilient networks by revealing the underlying mechanisms that reduce sensitivity to noise. In doing so, we introduce a new explainable regularization framework that harnesses these mechanisms to significantly enhance noise robustness in deep neural architectures.