.png){kind=icon}
返回
An Adaptive Robust Online Method for AUC Maximization
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2018.2869860.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Recently, increasing attention has been focused on the problem of online AUC maximization, and a great deal of efficient algorithms has been proposed. In spite of the promising performance of those online algorithms, however, most of them are sensitive to the outliers, which make them unsuitable for the applications with noisy data. To tackle the issue, in this paper, an adaptive robust method for online AUC maximization, termed AROAM is suggested. Specifically, a ramp loss based objective function oriented to AUC metric is firstly defined in AROAM, which has the strong ability of suppressing the influence of outliers. Then, concave-convex procedure is adopted for the convex approximation of the objective function. Finally, to further improve the performance of AROAM, an adaptive learning rate strategy is developed in each iteration, which can update the classifier effectively. Empirical studies on the benchmark data sets demonstrate the superiority of the proposed method in comparison with the state-of-the-arts.
Keyword:
AUC maximization algorithm
online learning
ramp loss
robust algorithm
adaptive learning rate
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3.6
论文数:
9.8W
被引数:
29.4W
机构
引用论文
分享收藏
分享收藏分享收藏Ramp loss one-class support vector machine; A robust and effective approach to anomaly detection problems斜坡损失一类支持向量机; 异常检测问题的鲁棒有效方法.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏
查看更多内容