arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1902.02771cs.CVcs.LGcs.NEeess.IV
收藏

全连通层对图像分类卷积神经网络性能的影响

Impact of Fully Connected Layers on Performance of Convolutional Neural Networks for Image Classification

Indian Institute of Information Technology Sri City(斯里城印度信息技术学院)

S. H. Shabbeer Basha, Shiv Ram Dubey, Viswanath Pulabaigari, Snehasis Mukherjee

摘要:

卷积神经网络(CNN)在计算机视觉等领域中,由于其能够从原始输入数据中学习特定于问题的特征,因此主要减少了对手工特征的需求。然而,选择特定于网络的CNN架构,这主要是通过经验或专业知识来执行的,是一个耗时且容易出错的过程。为了自动化学习CNN架构的过程,本文试图找到全连接(FC)层与数据集的一些特征之间的关系。CNN架构以及最近的数据集也被分类为深、浅、宽等。本文试图将这些术语形式化,并回答以下问题。(i)较深/较浅架构对CNN w.r.t.的性能有何影响?FC层?,(ii)更深/更宽的数据集如何影响CNN w.r.t. FC层?,以及(iii)哪种架构(较深/较浅)更适合于哪种(较深/较宽)数据集。为了解决这些发现,我们用三种具有不同深度的CNN架构进行了实验。通过改变FC层的数目进行实验。我们使用了四个广泛使用的数据集,包括CIFAR-10,CIFAR-100,Tiny ImageNet和CRCHistoPhenotypes,以证明我们在图像分类问题中的发现。这项研究的源代码可在https://github.com/shabbeersh/Impact-of-FC-layers上获得。

英文摘要:

The Convolutional Neural Networks (CNNs), in domains like computer vision, mostly reduced the need for handcrafted features due to its ability to learn the problem-specific features from the raw input data. However, the selection of dataset-specific CNN architecture, which mostly performed by either experience or expertise is a time-consuming and error-prone process. To automate the process of learning a CNN architecture, this paper attempts at finding the relationship between Fully Connected (FC) layers with some of the characteristics of the datasets. The CNN architectures, and recently datasets also, are categorized as deep, shallow, wide, etc. This paper tries to formalize these terms along with answering the following questions. (i) What is the impact of deeper/shallow architectures on the performance of the CNN w.r.t. FC layers?, (ii) How the deeper/wider datasets influence the performance of CNN w.r.t. FC layers?, and (iii) Which kind of architecture (deeper/ shallower) is better suitable for which kind of (deeper/ wider) datasets. To address these findings, we have performed experiments with three CNN architectures having different depths. The experiments are conducted by varying the number of FC layers. We used four widely used datasets including CIFAR-10, CIFAR-100, Tiny ImageNet, and CRCHistoPhenotypes to justify our findings in the context of the image classification problem. The source code of this research is available at https://github.com/shabbeersh/Impact-of-FC-layers.

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。

↑