当前位置:
X-MOL 学术
›
arXiv.cs.NE
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
MUXConv: Information Multiplexing in Convolutional Neural Networks
arXiv - CS - Neural and Evolutionary Computing Pub Date : 2020-03-31 , DOI: arxiv-2003.13880 Zhichao Lu and Kalyanmoy Deb and Vishnu Naresh Boddeti
arXiv - CS - Neural and Evolutionary Computing Pub Date : 2020-03-31 , DOI: arxiv-2003.13880 Zhichao Lu and Kalyanmoy Deb and Vishnu Naresh Boddeti
Convolutional neural networks have witnessed remarkable improvements in
computational efficiency in recent years. A key driving force has been the idea
of trading-off model expressivity and efficiency through a combination of
$1\times 1$ and depth-wise separable convolutions in lieu of a standard
convolutional layer. The price of the efficiency, however, is the sub-optimal
flow of information across space and channels in the network. To overcome this
limitation, we present MUXConv, a layer that is designed to increase the flow
of information by progressively multiplexing channel and spatial information in
the network, while mitigating computational complexity. Furthermore, to
demonstrate the effectiveness of MUXConv, we integrate it within an efficient
multi-objective evolutionary algorithm to search for the optimal model
hyper-parameters while simultaneously optimizing accuracy, compactness, and
computational efficiency. On ImageNet, the resulting models, dubbed MUXNets,
match the performance (75.3% top-1 accuracy) and multiply-add operations (218M)
of MobileNetV3 while being 1.6$\times$ more compact, and outperform other
mobile models in all the three criteria. MUXNet also performs well under
transfer learning and when adapted to object detection. On the ChestX-Ray 14
benchmark, its accuracy is comparable to the state-of-the-art while being
$3.3\times$ more compact and $14\times$ more efficient. Similarly, detection on
PASCAL VOC 2007 is 1.2% more accurate, 28% faster and 6% more compact compared
to MobileNetV2. Code is available from
https://github.com/human-analysis/MUXConv
中文翻译:
MUXConv:卷积神经网络中的信息复用
近年来,卷积神经网络在计算效率方面取得了显着进步。一个关键的驱动力是通过组合 $1\times 1$ 和深度可分离卷积代替标准卷积层来权衡模型表达能力和效率的想法。然而,效率的代价是网络中跨空间和通道的次优信息流。为了克服这一限制,我们提出了 MUXConv,该层旨在通过逐步复用网络中的通道和空间信息来增加信息流,同时降低计算复杂度。此外,为了证明 MUXConv 的有效性,我们将其集成到一个高效的多目标进化算法中,以搜索最佳模型超参数,同时优化准确性、紧凑性和计算效率。在 ImageNet 上,生成的模型被称为 MUXNets,与 MobileNetV3 的性能(75.3% top-1 准确率)和乘加运算(218M)相匹配,同时紧凑 1.6$\times$,并且在所有三个方面都优于其他移动模型标准。MUXNet 在迁移学习和适应对象检测时也表现良好。在 ChestX-Ray 14 基准测试中,它的准确性可与最先进的技术相媲美,同时更紧凑 3.3 美元,效率高 14 美元。同样,与 MobileNetV2 相比,在 PASCAL VOC 2007 上的检测准确度提高了 1.2%,速度提高了 28%,紧凑性提高了 6%。代码可从 https://github 获得。
更新日期:2020-04-08
中文翻译:
MUXConv:卷积神经网络中的信息复用
近年来,卷积神经网络在计算效率方面取得了显着进步。一个关键的驱动力是通过组合 $1\times 1$ 和深度可分离卷积代替标准卷积层来权衡模型表达能力和效率的想法。然而,效率的代价是网络中跨空间和通道的次优信息流。为了克服这一限制,我们提出了 MUXConv,该层旨在通过逐步复用网络中的通道和空间信息来增加信息流,同时降低计算复杂度。此外,为了证明 MUXConv 的有效性,我们将其集成到一个高效的多目标进化算法中,以搜索最佳模型超参数,同时优化准确性、紧凑性和计算效率。在 ImageNet 上,生成的模型被称为 MUXNets,与 MobileNetV3 的性能(75.3% top-1 准确率)和乘加运算(218M)相匹配,同时紧凑 1.6$\times$,并且在所有三个方面都优于其他移动模型标准。MUXNet 在迁移学习和适应对象检测时也表现良好。在 ChestX-Ray 14 基准测试中,它的准确性可与最先进的技术相媲美,同时更紧凑 3.3 美元,效率高 14 美元。同样,与 MobileNetV2 相比,在 PASCAL VOC 2007 上的检测准确度提高了 1.2%,速度提高了 28%,紧凑性提高了 6%。代码可从 https://github 获得。