Name
Title | ||
|---|---|---|
Managing the computational cost of model selection and cross-validation in extreme learning machines via Cholesky, SVD, QR and eigen decompositions. |
Abstract | ||
|---|---|---|
•Managing the computational cost of Extreme Learning Machine model selection and cross-validation.•Comparing Singular Value Decomposition (SVD), Eigenvalue Decomposition (EVD), Cholesky decomposition and QR decomposition.•Demonstrates theoretically and experimentally that matrix decompositions and cross-validation strategies play equally important roles in saving computational time.•Presents a fast and scalable 10-fold cross-validation version with Eigenvalue Decomposition. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1016/j.neucom.2018.01.005 | Neurocomputing |
Keywords | Field | DocType |
Model selection,Cross-validation,Extreme learning machines,Matrix decompositions | Singular value decomposition,Pattern recognition,Matrix (mathematics),Extreme learning machine,Matrix decomposition,Algorithm,Eigendecomposition of a matrix,Artificial intelligence,Triangular matrix,Mathematics,QR decomposition,Cholesky decomposition | Journal |
Volume | ISSN | Citations |
295 | 0925-2312 | 4 |
PageRank | References | Authors |
0.40 | 19 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yiannis Kokkinos | 1 | 33 | 6.56 |
| Konstantinos G. Margaritis | 2 | 303 | 45.46 |