Probabilistic Regularized Extreme Learning Machine for Robust Modeling of Noise Data

Probabilistic Regularized Extreme Learning Machine for Robust Modeling of Noise Data
复制标题

用于噪声数据鲁棒建模的概率正则极限学习机

DOI:
10.1109/tcyb.2017.2738060
复制
发表时间:
2018-08
影响因子:
11.8
通讯作者:
Li Han-Xiong
Li Han-Xiong
中科院分区:
计算机科学1区
文献类型:
--
作者:
Lu Xinjiang;Ming Li;Liu Wenbo;Li Han-Xiong

文献摘要

参考文献

被引文献

相似文献

极限学习机(ELM)在机器学习领域得到了广泛的研究,并因其简化的算法和降低的计算成本而被广泛实现。然而,它是不太有效的建模数据与非高斯噪声或数据包含离群值。在这里,提出了一种概率正则化ELM,以提高建模性能的数据包含非高斯噪声和/或离群值。传统的ELM通过最坏情况原理最小化建模误差,而所提出的方法构造了一个新的目标函数,以最小化该建模误差的均值和方差。因此,所提出的方法考虑建模误差分布。一个解决方案的方法,然后开发这个新的目标函数和所提出的方法被进一步证明是更强大的,与传统的ELM相比,即使受到噪声或离群值。实验结果表明,该方法对非高斯噪声或离群点问题具有较好的建模性能。
The extreme learning machine (ELM) has been extensively studied in the machine learning field and has been widely implemented due to its simplified algorithm and reduced computational costs. However, it is less effective for modeling data with non-Gaussian noise or data containing outliers. Here, a probabilistic regularized ELM is proposed to improve modeling performance with data containing non-Gaussian noise and/or outliers. While traditional ELM minimizes modeling error by using a worst-case scenario principle, the proposed method constructs a new objective function to minimize both mean and variance of this modeling error. Thus, the proposed method considers the modeling error distribution. A solution method is then developed for this new objective function and the proposed method is further proved to be more robust when compared with traditional ELM, even when subject to noise or outliers. Several experimental cases demonstrate that the proposed method has better modeling performance for problems with non-Gaussian noise or outliers.
用于短期负荷预测的切换延迟 PSO 优化极限学习机
DOI: 10.1016/j.neucom.2017.01.090
发表时间: 2017-05
期刊: Neurocomputing
影响因子: 6
作者:
Nianyin Zeng;Hong Zhang;Weibo Liu;Jinling Liang;Fuad E. Alsaadi
通讯作者: Fuad E. Alsaadi
DOI: 10.1109/tii.2015.2479852
发表时间: 2015-09
影响因子: 12.3
作者:
Xinjiang Lu;Chang Liu;Minghui Huang
通讯作者: Xinjiang Lu;Chang Liu;Minghui Huang
DOI: 10.1007/s13042-014-0292-7
发表时间: 2014-08
影响因子: 5.6
作者:
Huang, Yihua;Meng, Lei;Gong, Siyuan;Zhang, Guopeng
通讯作者: Zhang, Guopeng
用于回归问题的异常鲁棒极限学习机
DOI: 10.1016/j.neucom.2014.09.022
发表时间: 2015-03-03
期刊: NEUROCOMPUTING
影响因子: 6
作者:
Zhang, Kai;Luo, Minxia
通讯作者: Luo, Minxia
DOI: 10.4028/www.scientific.net/amr.989-994.3679
发表时间: 2014-07
期刊: Advanced Materials Research
影响因子: --
作者:
M. Ma;Bo He
通讯作者: M. Ma;Bo He