IMEXnet: A Forward Stable Deep Neural Network
IMEXnet: A Forward Stable Deep Neural Network
复制标题
DOI：
--
复制
发表时间：
2019-03
期刊：
影响因子：
--
通讯作者：
E. Haber;Keegan Lensink;Eran Treister;Lars Ruthotto
中科院分区：
文献类型：
--
作者：
E. Haber;Keegan Lensink;Eran Treister;Lars Ruthotto
深度卷积神经网络使许多机器学习和计算机视觉任务发生了革命性的变化，然而，一些仍然存在的关键挑战限制了它们的广泛使用。这些挑战包括提高网络对输入图像的扰动和卷积算子有限的“视场”的稳健性。我们介绍了IMEXnet，它通过采用偏微分方程组的半隐式方法来解决这些挑战。与类似的显式网络(如残差网络)相比，我们的网络更加稳定，最近的研究表明，它降低了对输入特征微小变化的敏感性，并提高了泛化能力。隐式步骤的添加连接了图像的每个通道中的所有像素，因此解决了视场问题，同时在参数数量和计算复杂性方面仍与标准卷积相当。我们还提出了一个新的用于语义分割的数据集，并使用NYU深度数据集演示了我们的体系结构的有效性。
Deep convolutional neural networks have revolutionized many machine learning and computer vision tasks, however, some remaining key challenges limit their wider use. These challenges include improving the network's robustness to perturbations of the input image and the limited ``field of view'' of convolution operators. We introduce the IMEXnet that addresses these challenges by adapting semi-implicit methods for partial differential equations. Compared to similar explicit networks, such as residual networks, our network is more stable, which has recently shown to reduce the sensitivity to small changes in the input features and improve generalization. The addition of an implicit step connects all pixels in each channel of the image and therefore addresses the field of view problem while still being comparable to standard convolutions in terms of the number of parameters and computational complexity. We also present a new dataset for semantic segmentation and demonstrate the effectiveness of our architecture using the NYU Depth dataset.