arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1703.07115cs.LG
收藏

使用内核相似性对深度网络进行分层训练

Layer-wise training of deep networks using kernel similarity

TCS Innovation Labs(TCS创新实验室)

Mandar Kulkarni, Shirish Karande

摘要:

深度学习已经在许多机器学习应用中显示出有希望的结果。由深度网络构建的分层特征表示可以实现数据的紧凑和精确编码。对训练后的深度网络的内核分析表明,使用更深的层,可以获得更简单、更准确的数据表示。在本文中,我们提出了一种用于监督分类任务的深度网络分层训练方法。每层的变换矩阵通过求解旨在更好的表示的优化来获得,其中后续层在由前一层产生的特征的顶部上构建其表示。我们将我们的方法的性能与使用反向传播训练的DNN进行了比较,后者具有与我们相同的架构。在真实图像数据集上的实验结果证明了该方法的有效性。我们还对层表示进行了内核分析,以验证更好的特征编码的说法。

英文摘要:

Deep learning has shown promising results in many machine learning applications. The hierarchical feature representation built by deep networks enable compact and precise encoding of the data. A kernel analysis of the trained deep networks demonstrated that with deeper layers, more simple and more accurate data representations are obtained. In this paper, we propose an approach for layer-wise training of a deep network for the supervised classification task. A transformation matrix of each layer is obtained by solving an optimization aimed at a better representation where a subsequent layer builds its representation on the top of the features produced by a previous layer. We compared the performance of our approach with a DNN trained using back-propagation which has same architecture as ours. Experimental results on the real image datasets demonstrate efficacy of our approach. We also performed kernel analysis of layer representations to validate the claim of better feature encoding.

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。

↑