arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1912.00664cs.CVcs.LG
收藏

利用响应对输入数据失真的统计依赖性训练卷积神经网络

Training the Convolutional Neural Network with Statistical Dependence of the Response on the Input Data Distortion

Federal Research Center Computer Science and Control of Russian Academy of Sciences(俄罗斯科学院计算机科学与控制联邦研究中心);Smart Engines Service LLC(Smart Engines服务有限责任公司);Institute for Information Transmission Problems of Russian Academy of Sciences(俄罗斯科学院信息传输问题研究所);Moscow Institute of Physics and Technology (State University)(莫斯科物理技术学院(国立大学))

Igor Janiszewski, Dmitry Slugin, Vladimir V. Arlazarov

摘要:

本文提出了一种利用输入数据的失真程度信息训练卷积神经网络的方法。学习过程被修改为一个额外的层,随后被删除,因此原始网络的架构不会改变。作为示例,考虑具有基于MNIST符号的训练数据和作为具有可变失真水平的高斯模糊的失真模型的LeNet5架构网络。这种方法没有网络的质量损失,并且在对测试数据的响应中具有显著的无差错区,这在传统的训练方法中是不存在的。这些响应在统计上依赖于输入图像的失真水平,并且它们之间存在很强的关系。

英文摘要:

The paper proposes an approach to training a convolutional neural network using information on the level of distortion of input data. The learning process is modified with an additional layer, which is subsequently deleted, so the architecture of the original network does not change. As an example, the LeNet5 architecture network with training data based on the MNIST symbols and a distortion model as Gaussian blur with a variable level of distortion is considered. This approach does not have quality loss of the network and has a significant error-free zone in responses on the test data which is absent in the traditional approach to training. The responses are statistically dependent on the level of input image's distortions and there is a presence of a strong relationship between them.

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。

↑