Direct kernel perceptron (DKP): ultra-fast kernel ELM-based classification with non-iterative closed-form weight calculation

DOI10.1016/J.NEUNET.2013.11.002MaRDI QIDQ470174zbMATHOpenAlexWikidataFDO

Authors Manuel Fernández-Delgado, Eva Cernadas, Senén Barro, Jorge Ribeiro, José Neves

Publication date 12 November 2014

Published in Neural Networks (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.neunet.2013.11.002

zbMATH Keywords

support vector machineclosed-form solutionextreme learning machineanalytical weight calculationkernel-based classificationmargin maximizationparallel Delta rule

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• A fast reduced kernel extreme learning machine
• DirectSVM: A simple support vector machine perceptron
• Parsimonious kernel extreme learning machine in primal via Cholesky factorization
• Distance based multiple kernel ELM: a fast multiple kernel learning approach
• 10.1162/153244303765208403

Cites work

• scientific article; zbMATH DE number 51537 (Why is no real title available?)
• A learning rule for very simple universal approximators consisting of a single layer of perceptrons
• Bagging predictors
• Random forests
• Statistical Analysis of Financial Data in S-Plus
• Support-vector networks

Cited in

(5)

• Grid topologies for the self-organizing map
• Two fast and accurate heuristic RBF learning rules for data classification
• An evaluation of the bihyperbolic function in the optimization of the backpropagation algorithm
• A fast SVD-hidden-nodes based extreme learning machine for large-scale data analytics
• Trends in extreme learning machines: a review

Describes a project that uses

Uses Software

• Matlab
• R
• UCI
• MASS (R)
• LIBSVM
• AdaBoost.MH
• SoftDoubleMaxMinOver

This page was built for publication: Direct kernel perceptron (DKP): ultra-fast kernel ELM-based classification with non-iterative closed-form weight calculation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q470174)