卷积神经网络
计算机科学
核(代数)
稳健性(进化)
卷积(计算机科学)
建筑
块(置换群论)
计算
背景(考古学)
人工智能
并行计算
计算机工程
算法
理论计算机科学
人工神经网络
数学
几何学
离散数学
艺术
古生物学
生物化学
化学
生物
视觉艺术
基因
作者
Xiaohan Ding,Yuchen Guo,Guiguang Ding,Jungong Han
出处
期刊:
[Institute of Electrical and Electronics Engineers]
日期:2019-10-01
卷期号:: 1911-1920
被引量:1192
标识
DOI:10.1109/iccv.2019.00200
摘要
As designing appropriate Convolutional Neural Network (CNN) architecture in the context of a given application usually involves heavy human works or numerous GPU hours, the research community is soliciting the architecture-neutral CNN structures, which can be easily plugged into multiple mature architectures to improve the performance on our real-world applications. We propose Asymmetric Convolution Block (ACB), an architecture-neutral structure as a CNN building block, which uses 1D asymmetric convolutions to strengthen the square convolution kernels. For an off-the-shelf architecture, we replace the standard square-kernel convolutional layers with ACBs to construct an Asymmetric Convolutional Network (ACNet), which can be trained to reach a higher level of accuracy. After training, we equivalently convert the ACNet into the same original architecture, thus requiring no extra computations anymore. We have observed that ACNet can improve the performance of various models on CIFAR and ImageNet by a clear margin. Through further experiments, we attribute the effectiveness of ACB to its capability of enhancing the model's robustness to rotational distortions and strengthening the central skeleton parts of square convolution kernels.
科研通智能强力驱动
Strongly Powered by AbleSci AI