Efficient Multiple Kernel Learning Algorithms Using Low-Rank Representation
| Autor: | Jianchuan Bai, Kewen Xia, Wenjia Niu, Baokai Zu |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
Support Vector Machine
General Computer Science Rank (linear algebra) Article Subject Computer science General Mathematics MathematicsofComputing_NUMERICALANALYSIS Datasets as Topic 02 engineering and technology Software_PROGRAMMINGTECHNIQUES lcsh:Computer applications to medicine. Medical informatics Machine learning computer.software_genre lcsh:RC321-571 Dimension (vector space) ComputingMethodologies_SYMBOLICANDALGEBRAICMANIPULATION 0202 electrical engineering electronic engineering information engineering Representation (mathematics) lcsh:Neurosciences. Biological psychiatry. Neuropsychiatry Multiple kernel learning business.industry General Neuroscience 020207 software engineering Pattern recognition General Medicine Constraint (information theory) Support vector machine ComputingMethodologies_PATTERNRECOGNITION Kernel (statistics) lcsh:R858-859.7 020201 artificial intelligence & image processing Pairwise comparison Artificial intelligence business computer Algorithm Algorithms Research Article |
| Zdroj: | Computational Intelligence and Neuroscience, Vol 2017 (2017) Computational Intelligence and Neuroscience |
| ISSN: | 1687-5265 |
| DOI: | 10.1155/2017/3678487 |
| Popis: | Unlike Support Vector Machine (SVM), Multiple Kernel Learning (MKL) allows datasets to be free to choose the useful kernels based on their distribution characteristics rather than a precise one. It has been shown in the literature that MKL holds superior recognition accuracy compared with SVM, however, at the expense of time consuming computations. This creates analytical and computational difficulties in solving MKL algorithms. To overcome this issue, we first develop a novel kernel approximation approach for MKL and then propose an efficient Low-Rank MKL (LR-MKL) algorithm by using the Low-Rank Representation (LRR). It is well-acknowledged that LRR can reduce dimension while retaining the data features under a global low-rank constraint. Furthermore, we redesign the binary-class MKL as the multiclass MKL based on pairwise strategy. Finally, the recognition effect and efficiency of LR-MKL are verified on the datasets Yale, ORL, LSVT, and Digit. Experimental results show that the proposed LR-MKL algorithm is an efficient kernel weights allocation method in MKL and boosts the performance of MKL largely. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|