当前位置: X-MOL 学术IEEE Trans. Cybern. › 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.)
Large-Scale Robust Semisupervised Classification
IEEE Transactions on Cybernetics ( IF 11.8 ) Pub Date : 2019-03-01 , DOI: 10.1109/tcyb.2018.2789420
Lingling Zhang , Minnan Luo , Zhihui Li , Feiping Nie , Huaxiang Zhang , Jun Liu , Qinghua Zheng

Semisupervised learning aims to leverage both labeled and unlabeled data to improve performance, where most of them are graph-based methods. However, the graph-based semisupervised methods are not capable for large-scale data since the computational consumption on the construction of graph Laplacian matrix is huge. On the other hand, the substantial unlabeled data in training stage of semisupervised learning could cause large uncertainties and potential threats. Therefore, it is crucial to enhance the robustness of semisupervised classification. In this paper, a novel large-scale robust semisupervised learning method is proposed in the framework of capped $\ell _{\boldsymbol {2,p}}$ -norm. This strategy is superior not only in computational cost because it makes the graph Laplacian matrix unnecessary, but also in robustness to outliers since the capped $\ell _{\boldsymbol {2,p}}$ -norm used for loss measurement. An efficient optimization algorithm is exploited to solve the nonconvex and nonsmooth challenging problem. The complexity of the proposed algorithm is analyzed and discussed in theory detailedly. Finally, extensive experiments are conducted over six benchmark data sets to demonstrate the effectiveness and superiority of the proposed method.

中文翻译:

大规模鲁棒半监督分类

半监督学习旨在利用标记和未标记的数据来提高性能,其中大多数都是基于图的方法。但是,基于图的半监督方法无法处理大规模数据,因为构建图拉普拉斯矩阵的计算量很大。另一方面,在半监督学习的训练阶段大量未标记的数据可能会导致很大的不确定性和潜在的威胁。因此,提高半监督分类的鲁棒性至关重要。本文在有上限的框架下,提出了一种新颖的大规模鲁棒的半监督学习方法。 $ \ ell _ {\ boldsymbol {2,p}} $ -规范。这种策略不仅在计算成本上是优越的,因为它使图Laplacian矩阵成为不必要,而且在对异常值进行自定义的鲁棒性之后,其鲁棒性也得到了提高。 $ \ ell _ {\ boldsymbol {2,p}} $ -范数用于损耗测量。开发了一种有效的优化算法来解决非凸且非平滑的挑战性问题。理论上详细分析和讨论了所提出算法的复杂性。最后,在六个基准数据集上进行了广泛的实验,以证明该方法的有效性和优越性。
更新日期:2019-03-01
down
wechat
bug