4.7 Article

COMO: Efficient Deep Neural Networks Expansion With COnvolutional MaxOut

期刊

IEEE TRANSACTIONS ON MULTIMEDIA
卷 23, 期 -, 页码 1722-1730

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMM.2020.3002614

关键词

Convolution; Spatial resolution; Convolutional neural networks; Deep learning; Computer architecture; Transforms; Artificial neural networks; computational and artificial intelligence; feedforward neural networks; multilayer perceptrons; neural networks

资金

  1. National Key R&D Program of China [2018YFB1402600, 2019YFB2102100]
  2. Shenzhen Discipline Construction Project for Urban Computing and Data Intelligence
  3. Shenzhen Engineering Research Center for Beidou Positioning Service Improvement Technology [XMHT20190101035]

向作者/读者索取更多资源

COMO is a new deep convolutional neural network extension strategy that improves classification accuracy without increasing parameter size through a split-transform-merge paradigm. Compared to existing solutions, it performs well across various architectures.
In this paper, we extend the classic MaxOut strategy, originally designed for Multiple Layer Preceptors (MLPs), into COnvolutional MaxOut (COMO) - a new strategy making deep convolutional neural networks wider with parameter efficiency. Compared to the existing solutions, such as ResNeXt for ResNet or Inception for VGG-alikes, COMO works well on both linear architectures and the ones with skipped connections and residual blocks. More specifically, COMO adopts a novel split-transform-merge paradigm that extends the layers with spatial resolution reduction into multiple parallel splits. For the layer with COMO, each split passes the input feature maps through a 4D convolution operator with independent batch normalization operators for transformation, then merge into the aggregated output of the original sizes through max-pooling. Such a strategy is expected to tackle the potential classification accuracy degradation due to the spatial resolution reduction, by incorporating the multiple splits and max-pooling-based feature selection. Our experiment using a wide range of deep architectures shows that COMO can significantly improve the classification accuracy of ResNet/VGG-alike networks based on a large number of benchmark datasets. COMO further outperforms the existing solutions, e.g., Inceptions, ResNeXts, SE-ResNet, and Xception, that make networks wider, and it dominates in the comparison of accuracy versus parameter sizes.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据