Parallel online sequential extreme learning machine based on MapReduce

Wang, Botao; Huang, Shongming; Qiu, Junhao; Liu, Yu; Wang, Guoren

doi:10.1016/j.neucom.2014.03.076

Cited by 98 publications

(53 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Wang et al [12] have made first efforts to implement parallel ELM using MapReduce on some classification problems. A parallel implementation of EELM or ML-EELM is recommended as possible future research.…”

Section: Resultsmentioning

confidence: 99%

Fast Handwritten Digit Recognition with Multilayer Ensemble Extreme Learning Machine

Noushahr

Ahmadi

Casey

2015

Research and Development in Intelligent Systems XXXII

View full text Add to dashboard Cite

Conventional artificial neural networks and convolutional neural networks perform well on the task of automatic handwriting recognition. But, they suffer from long training times and their complex nature. An alternative learning algorithm called Extreme Learning Machine overcomes these shortcomings by determining the weights of a neural network analytically. In this paper, a novel classifier based on Extreme Learning Machine is proposed that achieves competitive accuracy results while keeping training times low. This classifier is called multilayer ensemble Extreme Learning Machine. The novel classifier is evaluated against traditional backpropagation and Extreme Learning Machine on the well-known MNIST dataset. Possible future work on parallel Extreme Learning Machine is shown up.

show abstract

Section: Resultsmentioning

confidence: 99%

Fast Handwritten Digit Recognition with Multilayer Ensemble Extreme Learning Machine

Noushahr

Ahmadi

Casey

2015

Research and Development in Intelligent Systems XXXII

View full text Add to dashboard Cite

show abstract

“…With the fast developments in big data and distributed systems, there are also literatures dedicated to making the algorithm adaptive to large scale datasets [42] or Map-Reduce framework [43,44].…”

Section: Extreme Learning Machinementioning

confidence: 99%

Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E‐Nose Systems

Ma¹,

Luo²,

Qin³

et al. 2018

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Machine learning approaches have been widely used to tackle the problem of sensor array drift in E-Nose systems. However, labeled data are rare in practice, which makes supervised learning methods hard to be applied. Meanwhile, current solutions require updating the analytical model in an offline manner, which hampers their uses for online scenarios. In this paper, we extended Target Domain Adaptation Extreme Learning Machine (DAELM T) to achieve high accuracy with less labeled samples by proposing a Weighted Domain Transfer Extreme Learning Machine, which uses clustering information as prior knowledge to help select proper labeled samples and calculate sensitive matrix for weighted learning. Furthermore, we converted DAELM T and the proposed method into their online learning versions under which scenario the labeled data are selected beforehand. Experimental results show that, for batch learning version, the proposed method uses around 20% less labeled samples while achieving approximately equivalent or better accuracy. As for the online versions, the methods maintain almost the same accuracies as their offline counterparts do, but the time cost remains around a constant value while that of offline versions grows with the number of samples.

show abstract

“…Load forecasting problems, involving intelligent ML solutions pertaining to large volume of data source generation, are a perfect fit for MapReduce deployments [76].…”

Section: Mapreduce Parallel Processingmentioning

confidence: 99%

Big Data Analytics for Dynamic Energy Management in Smart Grids

2015

View full text Add to dashboard Cite

The smart electricity grid enables a two-way flow of power and data between suppliers and consumers in order to facilitate the power flow optimization in terms of economic efficiency, reliability and sustainability. This infrastructure permits the consumers and the micro-energy producers to take a more active role in the electricity market and the dynamic energy management (DEM). The most important challenge in a smart grid (SG) is how to take advantage of the users' participation in order to reduce the cost of power. However, effective DEM depends critically on load and renewable production forecasting. This calls for intelligent methods and solutions for the real-time exploitation of the large volumes of data generated by a vast amount of smart meters. Hence, robust data analytics, high performance computing, efficient data network management, and cloud computing techniques are critical towards the optimized operation of SGs. This research aims to highlight the big data issues and challenges faced by the DEM employed in SG networks. It also provides a brief description of the most commonly used data processing methods in the literature, and proposes a promising direction for future research in the field.Comment: Published in ELSEVIER Big Data Researc

show abstract

Parallel online sequential extreme learning machine based on MapReduce

Cited by 98 publications

References 8 publications

Fast Handwritten Digit Recognition with Multilayer Ensemble Extreme Learning Machine

Fast Handwritten Digit Recognition with Multilayer Ensemble Extreme Learning Machine

Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E‐Nose Systems

Big Data Analytics for Dynamic Energy Management in Smart Grids

Contact Info

Product

Resources

About