论文标题
用于预测弹性和恢复神经网络损害的几何算法
Geometric algorithms for predicting resilience and recovering damage in neural networks
论文作者
论文摘要
尽管电路损坏很大，但生物神经网络仍以维持性能的发展。为了生存损害，生物网络体系结构既具有对组件损失的固有弹性，又具有通过可塑性来调节网络权重以稳定性能的恢复程序。尽管在技术应用中具有弹性的重要性，但人工神经网络的弹性知之甚少，并且尚未开发自主恢复算法。在本文中，我们建立了一个数学框架，以通过差异几何形状的镜头来分析人工神经网络的弹性。我们的几何语言提供了自然算法，可确定训练有素的网络中的本地漏洞以及动态调整网络以补偿损害的恢复算法。我们在常用图像分析网络中揭示了惊人的漏洞，例如MLP和CNN对MNIST和CIFAR10进行了训练。我们还发现了高性能恢复路径，使同一网络能够动态重新调整其参数以补偿损害。从广义上讲，我们的工作提供了使人工系统具有韧性和快速恢复程序的程序，以增强其与物联网设备的集成，并使他们在关键应用程序中的部署。
Biological neural networks have evolved to maintain performance despite significant circuit damage. To survive damage, biological network architectures have both intrinsic resilience to component loss and also activate recovery programs that adjust network weights through plasticity to stabilize performance. Despite the importance of resilience in technology applications, the resilience of artificial neural networks is poorly understood, and autonomous recovery algorithms have yet to be developed. In this paper, we establish a mathematical framework to analyze the resilience of artificial neural networks through the lens of differential geometry. Our geometric language provides natural algorithms that identify local vulnerabilities in trained networks as well as recovery algorithms that dynamically adjust networks to compensate for damage. We reveal striking vulnerabilities in commonly used image analysis networks, like MLP's and CNN's trained on MNIST and CIFAR10 respectively. We also uncover high-performance recovery paths that enable the same networks to dynamically re-adjust their parameters to compensate for damage. Broadly, our work provides procedures that endow artificial systems with resilience and rapid-recovery routines to enhance their integration with IoT devices as well as enable their deployment for critical applications.
