Outlier-robust extreme learning machine for regression problems
Outlier-robust extreme learning machine for regression problems
复制标题
用于回归问题的异常鲁棒极限学习机
DOI：
10.1016/j.neucom.2014.09.022
复制
发表时间：
2015-03-03
期刊：
影响因子：
6
通讯作者：
Luo, Minxia
中科院分区：
文献类型：
--
作者：
Zhang, Kai;Luo, Minxia
极限学习机作为机器学习中最有用的技术之一，因其独特的超快速学习能力而受到广泛关注。特别是，它被广泛认为是ELM具有速度优势，同时执行令人满意的结果。然而，异常值的存在可能会导致ELM模型不可靠。在本文中，我们的研究解决了ELM在回归问题中的离群鲁棒性。基于离群点的稀疏特性，提出了一种离群鲁棒ELM，其中使用l（1）范数损失函数来增强鲁棒性。特别是，快速和准确的增广拉格朗日乘子方法，以保证有效性和效率。通过函数逼近实验和实际应用表明，该方法不仅保持了ELM的优点，而且在处理异常数据时具有显著的稳定精度。(C)2014爱思唯尔有限公司版权所有。
Extreme learning machine (ELM), as one of the most useful techniques in machine learning, has attracted extensive attentions due to its unique ability for extremely fast learning. In particular, it is widely recognized that ELM has speed advantage while performing satisfying results. However, the presence of outliers may give rise to unreliable ELM model. In this paper, our study addresses the outlier robustness of ELM in regression problems. Based on the sparsity characteristic of outliers, this work proposes an outlier-robust ELM where the l(1)-norm loss function is used to enhance the robustness. Specially, the fast and accurate augmented Iagrangian multiplier method is applied to guarantee the effectiveness and efficiency. According to the experiments on function approximation and some real-world applications, the proposed approach not only maintains the advantages from original ELM, but also shows notable and stable accuracy in handling data with outliers. (C) 2014 Elsevier B.V. All rights reserved.