理解深度卷积网络。
Understanding deep convolutional networks.
作者信息
Mallat Stéphane
机构信息
École Normale Supérieure, CNRS, PSL, 45 rue d'Ulm, Paris, France
出版信息
Philos Trans A Math Phys Eng Sci. 2016 Apr 13;374(2065):20150203. doi: 10.1098/rsta.2015.0203.
Abstract
Deep convolutional networks provide state-of-the-art classifications and regressions results over many high-dimensional problems. We review their architecture, which scatters data with a cascade of linear filter weights and nonlinearities. A mathematical framework is introduced to analyse their properties. Computations of invariants involve multiscale contractions with wavelets, the linearization of hierarchical symmetries and sparse separations. Applications are discussed.
摘要
深度卷积网络在许多高维问题上提供了最先进的分类和回归结果。我们回顾了它们的架构，该架构通过一系列线性滤波器权重和非线性函数来分散数据。引入了一个数学框架来分析它们的属性。不变量的计算涉及小波的多尺度收缩、层次对称性的线性化和稀疏分离。文中还讨论了相关应用。
相似文献
1
Understanding deep convolutional networks.理解深度卷积网络。
Philos Trans A Math Phys Eng Sci. 2016 Apr 13;374(2065):20150203. doi: 10.1098/rsta.2015.0203.
2
Hierarchical Recurrent Neural Hashing for Image Retrieval With Hierarchical Convolutional Features.基于层次卷积特征的层次递归神经网络哈希图像检索
IEEE Trans Image Process. 2018;27(1):106-120. doi: 10.1109/TIP.2017.2755766.
3
Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology.基于数字病理切片全场图像的深度学习卷积神经网络用于胃癌的自动分类。
Comput Med Imaging Graph. 2017 Nov;61:2-13. doi: 10.1016/j.compmedimag.2017.06.001. Epub 2017 Jun 16.
4
A novel end-to-end classifier using domain transferred deep convolutional neural networks for biomedical images.一种使用域转移深度卷积神经网络的新型端到端生物医学图像分类器。
Comput Methods Programs Biomed. 2017 Mar;140:283-293. doi: 10.1016/j.cmpb.2016.12.019. Epub 2017 Jan 6.
5
Towards dropout training for convolutional neural networks.面向卷积神经网络的随机失活训练
Neural Netw. 2015 Nov;71:1-10. doi: 10.1016/j.neunet.2015.07.007. Epub 2015 Jul 29.
6
Deep Convolutional Neural Networks for large-scale speech tasks.用于大规模语音任务的深度卷积神经网络。
Neural Netw. 2015 Apr;64:39-48. doi: 10.1016/j.neunet.2014.08.005. Epub 2014 Sep 16.
7
Sensor Fusion for Myoelectric Control Based on Deep Learning With Recurrent Convolutional Neural Networks.基于具有递归卷积神经网络的深度学习的肌电控制传感器融合
Artif Organs. 2018 Sep;42(9):E272-E282. doi: 10.1111/aor.13153. Epub 2018 Jul 13.
8
Brain tumor segmentation with Deep Neural Networks.基于深度神经网络的脑肿瘤分割。
Med Image Anal. 2017 Jan;35:18-31. doi: 10.1016/j.media.2016.05.004. Epub 2016 May 19.
9
Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection.基于极端学习机的卷积神经网络快速学习方法及其在车道检测中的应用。
Neural Netw. 2017 Mar;87:109-121. doi: 10.1016/j.neunet.2016.12.002. Epub 2016 Dec 10.
10
Stacked Convolutional Denoising Auto-Encoders for Feature Representation.堆叠卷积去噪自编码器的特征表示。
IEEE Trans Cybern. 2017 Apr;47(4):1017-1027. doi: 10.1109/TCYB.2016.2536638. Epub 2016 Mar 16.
引用本文的文献
1
A model for tobacco growing area classification based on time series features of thermogravimetric analysis.一种基于热重分析时间序列特征的烟草种植面积分类模型。
Biotechnol Biofuels Bioprod. 2025 Aug 11;18(1):90. doi: 10.1186/s13068-025-02682-x.
2
A novel approach to graph distinction through GENEOs and permutants.一种通过基因（GENEOs）和置换体进行图形区分的新方法。
Sci Rep. 2025 Feb 20;15(1):6259. doi: 10.1038/s41598-025-90152-7.
3
Investigating the Effect of Vibration Signal Length on Bearing Fault Classification Using Wavelet Scattering Transform.研究振动信号长度对基于小波散射变换的轴承故障分类的影响。
Sensors (Basel). 2025 Jan 24;25(3):699. doi: 10.3390/s25030699.
4
From motion to emotion: exploring challenging behaviors in autism spectrum disorder through analysis of wearable physiology and movement.从运动到情感：通过可穿戴生理和运动分析探索自闭症谱系障碍中的挑战性行为。
Physiol Meas. 2025 Jan 29;13(1):015004. doi: 10.1088/1361-6579/ada51b.
5
Harnessing Deep Learning for Accurate Pathological Assessment of Brain Tumor Cell Types.利用深度学习进行脑肿瘤细胞类型的准确病理评估。
J Imaging Inform Med. 2025 Apr;38(2):1098-1111. doi: 10.1007/s10278-024-01107-9. Epub 2024 Aug 16.
6
Wavelet scattering networks in deep learning for discovering protein markers in a cohort of Swedish rectal cancer patients.深度学习中的小波散射网络在发现瑞典直肠癌患者队列中的蛋白质标记物中的应用。
Cancer Med. 2023 Dec;12(23):21502-21518. doi: 10.1002/cam4.6672. Epub 2023 Nov 28.
7
Functional Connectivity and Feature Fusion Enhance Multiclass Motor-Imagery Brain-Computer Interface Performance.功能连接和特征融合提高多类运动想象脑-机接口性能。
Sensors (Basel). 2023 Aug 30;23(17):7520. doi: 10.3390/s23177520.
8
Covariance properties under natural image transformations for the generalised Gaussian derivative model for visual receptive fields.视觉感受野广义高斯导数模型在自然图像变换下的协方差特性
Front Comput Neurosci. 2023 Jun 15;17:1189949. doi: 10.3389/fncom.2023.1189949. eCollection 2023.
9
Development of Machine-Learning Models for Tinnitus-Related Distress Classification Using Wavelet-Transformed Auditory Evoked Potential Signals and Clinical Data.使用小波变换听觉诱发电位信号和临床数据开发用于耳鸣相关痛苦分类的机器学习模型
J Clin Med. 2023 Jun 4;12(11):3843. doi: 10.3390/jcm12113843.
10
[Primary study on recognition of vascular stiffness based on wavelet scattering neural network].基于小波散射神经网络的血管硬度识别初步研究
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2023 Apr 25;40(2):244-248. doi: 10.7507/1001-5515.202207068.
本文引用的文献
1
Deep learning.深度学习。
Nature. 2015 May 28;521(7553):436-44. doi: 10.1038/nature14539.
2
Deep learning of the tissue-regulated splicing code.深度学习组织调控的剪接代码。
Bioinformatics. 2014 Jun 15;30(12):i121-9. doi: 10.1093/bioinformatics/btu277.
3
Invariant scattering convolution networks.不变散射卷积网络。
IEEE Trans Pattern Anal Mach Intell. 2013 Aug;35(8):1872-86. doi: 10.1109/TPAMI.2012.230.
Zotero 插件