Revisiting Sparse Convolutional Model for Visual Recognition
Revisiting Sparse Convolutional Model for Visual Recognition
复制标题
DOI：
10.48550/arxiv.2210.12945
复制
发表时间：
2022-10
期刊：
影响因子：
--
通讯作者：
Xili Dai;Mingyang Li;Pengyuan Zhai;Shengbang Tong;Xingjian Gao;Shao-Lun Huang;Zhihui Zhu;Chong You;Y. Ma
中科院分区：
文献类型：
--
作者：
Xili Dai;Mingyang Li;Pengyuan Zhai;Shengbang Tong;Xingjian Gao;Shao-Lun Huang;Zhihui Zhu;Chong You;Y. Ma
尽管深度神经网络在图像分类方面具有很强的经验性能，但它通常被视为“黑匣子”，难以解释。另一方面，稀疏卷积模型假设信号可以由卷积字典中的几个元素的线性组合表示，是分析自然图像的强大工具，具有良好的理论可解释性和生物可解释性。然而，与经验设计的深度网络相比，这种原则模型并没有表现出竞争力。本文重新审视了图像分类的稀疏卷积模型，并弥合了（深度学习的）良好经验性能和（稀疏卷积模型的）良好可解释性之间的差距。我们的方法使用从卷积稀疏编码定义的可微优化层作为传统深度神经网络中标准卷积层的直接替代。我们表明，与传统神经网络相比，这些模型在CIFAR-10，CIFAR-100和ImageNet数据集上具有同样强大的经验性能。通过利用稀疏建模的稳定恢复特性，我们进一步证明了这种模型可以通过稀疏正则化和数据重建项之间的简单适当权衡，对输入损坏以及测试中的对抗性扰动更加鲁棒。源代码可以在https://github.com/Delay-Xili/SDNet上找到。
Despite strong empirical performance for image classification, deep neural networks are often regarded as ``black boxes'' and they are difficult to interpret. On the other hand, sparse convolutional models, which assume that a signal can be expressed by a linear combination of a few elements from a convolutional dictionary, are powerful tools for analyzing natural images with good theoretical interpretability and biological plausibility. However, such principled models have not demonstrated competitive performance when compared with empirically designed deep networks. This paper revisits the sparse convolutional modeling for image classification and bridges the gap between good empirical performance (of deep learning) and good interpretability (of sparse convolutional models). Our method uses differentiable optimization layers that are defined from convolutional sparse coding as drop-in replacements of standard convolutional layers in conventional deep neural networks. We show that such models have equally strong empirical performance on CIFAR-10, CIFAR-100, and ImageNet datasets when compared to conventional neural networks. By leveraging stable recovery property of sparse modeling, we further show that such models can be much more robust to input corruptions as well as adversarial perturbations in testing through a simple proper trade-off between sparse regularization and data reconstruction terms. Source code can be found at https://github.com/Delay-Xili/SDNet.