Distributed Extreme Learning Machine for Nonlinear Learning over Network
Distributed data collection and analysis over a network are ubiquitous, especially over a wireless sensor network (WSN). To our knowledge, the data model used in most of the distributed algorithms is linear. However, in real applications, the linearity of systems is not always guaranteed. In nonline...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2015-02-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1099-4300/17/2/818 |
| _version_ | 1828395646086807552 |
|---|---|
| author | Songyan Huang Chunguang Li |
| author_facet | Songyan Huang Chunguang Li |
| author_sort | Songyan Huang |
| collection | DOAJ |
| description | Distributed data collection and analysis over a network are ubiquitous, especially over a wireless sensor network (WSN). To our knowledge, the data model used in most of the distributed algorithms is linear. However, in real applications, the linearity of systems is not always guaranteed. In nonlinear cases, the single hidden layer feedforward neural network (SLFN) with radial basis function (RBF) hidden neurons has the ability to approximate any continuous functions and, thus, may be used as the nonlinear learning system. However, confined by the communication cost, using the distributed version of the conventional algorithms to train the neural network directly is usually prohibited. Fortunately, based on the theorems provided in the extreme learning machine (ELM) literature, we only need to compute the output weights of the SLFN. Computing the output weights itself is a linear learning problem, although the input-output mapping of the overall SLFN is still nonlinear. Using the distributed algorithmto cooperatively compute the output weights of the SLFN, we obtain a distributed extreme learning machine (dELM) for nonlinear learning in this paper. This dELM is applied to the regression problem and classification problem to demonstrate its effectiveness and advantages. |
| first_indexed | 2024-12-10T08:16:42Z |
| format | Article |
| id | doaj.art-191df3204339463b923b6167bfbe50b1 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-12-10T08:16:42Z |
| publishDate | 2015-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-191df3204339463b923b6167bfbe50b12022-12-22T01:56:27ZengMDPI AGEntropy1099-43002015-02-0117281884010.3390/e17020818e17020818Distributed Extreme Learning Machine for Nonlinear Learning over NetworkSongyan Huang0Chunguang Li1Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, ChinaDepartment of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, ChinaDistributed data collection and analysis over a network are ubiquitous, especially over a wireless sensor network (WSN). To our knowledge, the data model used in most of the distributed algorithms is linear. However, in real applications, the linearity of systems is not always guaranteed. In nonlinear cases, the single hidden layer feedforward neural network (SLFN) with radial basis function (RBF) hidden neurons has the ability to approximate any continuous functions and, thus, may be used as the nonlinear learning system. However, confined by the communication cost, using the distributed version of the conventional algorithms to train the neural network directly is usually prohibited. Fortunately, based on the theorems provided in the extreme learning machine (ELM) literature, we only need to compute the output weights of the SLFN. Computing the output weights itself is a linear learning problem, although the input-output mapping of the overall SLFN is still nonlinear. Using the distributed algorithmto cooperatively compute the output weights of the SLFN, we obtain a distributed extreme learning machine (dELM) for nonlinear learning in this paper. This dELM is applied to the regression problem and classification problem to demonstrate its effectiveness and advantages.http://www.mdpi.com/1099-4300/17/2/818distributed learningextreme learning machinenonlinear learningdiffusionleast-mean square (LMS)recursive least squares (RLS)regressionclassification |
| spellingShingle | Songyan Huang Chunguang Li Distributed Extreme Learning Machine for Nonlinear Learning over Network Entropy distributed learning extreme learning machine nonlinear learning diffusion least-mean square (LMS) recursive least squares (RLS) regression classification |
| title | Distributed Extreme Learning Machine for Nonlinear Learning over Network |
| title_full | Distributed Extreme Learning Machine for Nonlinear Learning over Network |
| title_fullStr | Distributed Extreme Learning Machine for Nonlinear Learning over Network |
| title_full_unstemmed | Distributed Extreme Learning Machine for Nonlinear Learning over Network |
| title_short | Distributed Extreme Learning Machine for Nonlinear Learning over Network |
| title_sort | distributed extreme learning machine for nonlinear learning over network |
| topic | distributed learning extreme learning machine nonlinear learning diffusion least-mean square (LMS) recursive least squares (RLS) regression classification |
| url | http://www.mdpi.com/1099-4300/17/2/818 |
| work_keys_str_mv | AT songyanhuang distributedextremelearningmachinefornonlinearlearningovernetwork AT chunguangli distributedextremelearningmachinefornonlinearlearningovernetwork |