arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1803.04022cs.CV
收藏

深度词典学习:一种参数网络方法

Deep Dictionary Learning: A PARametric NETwork Approach

North Carolina State University(北卡罗来纳州立大学);Army Research Office(陆军研究办公室)

Shahin Mahdizadehaghdam, Ashkan Panahi, Hamid Krim, Liyi Dai

摘要:

深度字典学习在不同的图像尺度上寻找多个字典,以捕捉互补的连贯特征。我们提出了一种方法,学习层次的合成字典与图像分类的目标。通过分类目标训练字典和分类参数,并通过减少每层中的重建损失来提取稀疏特征。重建目标在某种意义上规范了分类问题,并在提取的特征中注入源信号信息。所提出的分层方法的性能通过增加更多的层来提高,从而使该模型更容易调整和适应。此外,该算法在对抗性扰动的存在下显示出显着较低的欺骗率。所提出的方法的验证是基于其分类性能,使用四个基准数据集,并比较类似大小的CNN。

英文摘要:

Deep dictionary learning seeks multiple dictionaries at different image scales to capture complementary coherent characteristics. We propose a method for learning a hierarchy of synthesis dictionaries with an image classification goal. The dictionaries and classification parameters are trained by a classification objective, and the sparse features are extracted by reducing a reconstruction loss in each layer. The reconstruction objectives in some sense regularize the classification problem and inject source signal information in the extracted features. The performance of the proposed hierarchical method increases by adding more layers, which consequently makes this model easier to tune and adapt. The proposed algorithm furthermore, shows remarkably lower fooling rate in presence of adversarial perturbation. The validation of the proposed approach is based on its classification performance using four benchmark datasets and is compared to a CNN of similar size.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑