arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1805.03444cs.LGcs.CRcs.CVstat.ML
收藏

利用卷积神经网络各层输入特征图控制隐私丢失

Controlling the privacy loss with the input feature maps of the layers in convolutional neural networks

Samsung Electronics(三星电子)

Woohyung Chun, Sung-Min Hong, Junho Huh, Inyup Kang

摘要:

我们提出了一种方法来净化输入特征映射(IFM)的隐私,这些IFM被馈送到CNN(卷积神经网络)的层。该方法引入了使使用CNN的应用能够控制隐私损失的净化程度,该隐私损失表示为原始IFM和净化后的IFM的概率准确度的比率。对于IFM的净化,设计了基于采样保持的近似方案以满足特定于应用的净化程度。该方案通过用最接近采样窗口均值的非零样本替换窗口中的所有样本来逼近IFM。它还通过将多维IFM张量展开为待近似的一维流来消除对CNN配置的依赖性。

英文摘要:

We propose the method to sanitize the privacy of the IFM(Input Feature Map)s that are fed into the layers of CNN(Convolutional Neural Network)s. The method introduces the degree of the sanitization that makes the application using a CNN be able to control the privacy loss represented as the ratio of the probabilistic accuracies for original IFM and sanitized IFM. For the sanitization of an IFM, the sample-and-hold based approximation scheme is devised to satisfy an application-specific degree of the sanitization. The scheme approximates an IFM by replacing all the samples in a window with the non-zero sample closest to the mean of the sampling window. It also removes the dependency on CNN configuration by unfolding multi-dimensional IFM tensors into one-dimensional streams to be approximated.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑