FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2263177918> ?p ?o ?g. }

• W2263177918 abstract "Kernel-based algorithms have been used with great success in a variety of machine learning applications. These include algorithms such as support vector machines for classification, kernel ridge regression, ranking algorithms, clustering algorithms, and virtually all popular dimensionality reduction algorithms. But, the choice of the kernel, which is crucial to the success of these algorithms, has been traditionally left entirely to the user. Rather than requesting the user to commit to a specific kernel, multiple kernel algorithms require the user only to specify a family of kernels. This family of kernels can be used by a learning algorithm to form a combined kernel and derive an accurate predictor. This is a problem that has attracted a lot of attention recently, both from the theoretical point of view and from the algorithmic, optimization, and application point of view. This thesis presents a number of novel theoretical and algorithmic results for learning with multiple kernels. It gives the first tight margin-based generalization bounds for learning kernels with Lp regularization. In particular, our margin bounds for L1 regularization are shown to have only a logarithmic dependency on the number of kernels, which is a significant improvement over all previous analyses. Our results also include stability-based guarantees for a class of regression algorithms. In all cases, these guarantees indicate the benefits of learning with a large number of kernels. We also present a family of new two-stage algorithms for learning kernels based on a notion of alignment and give an extensive analysis of the properties of these algorithms. We show the existence of good predictors for the notion of alignment we define and give efficient algorithms for learning a maximum alignment kernel by showing that the problem can be reduced to a simple quadratic program. Finally, we report the results of extensive experiments with our two-stage algorithms, which show an improvement both over the uniform combination of kernels and over other state-of-the-art learning kernel methods for L1 and L2 regularization. These might constitute the first series of results for learning with multiple kernels that demonstrate a consistent improvement over a uniform combination of kernels." @default.
• W2263177918 created "2016-06-24" @default.
• W2263177918 creator A5058849006 @default.
• W2263177918 creator A5089765894 @default.
• W2263177918 date "2010-01-01" @default.
• W2263177918 modified "2023-09-23" @default.
• W2263177918 title "Theoretical foundations and algorithms for learning with multiple kernels" @default.
• W2263177918 cites W1487813067 @default.
• W2263177918 cites W1500474910 @default.
• W2263177918 cites W1510073064 @default.
• W2263177918 cites W1518932425 @default.
• W2263177918 cites W1524645404 @default.
• W2263177918 cites W1542886316 @default.
• W2263177918 cites W1559159361 @default.
• W2263177918 cites W1564947197 @default.
• W2263177918 cites W1608733719 @default.
• W2263177918 cites W1646506067 @default.
• W2263177918 cites W1755117326 @default.
• W2263177918 cites W183163904 @default.
• W2263177918 cites W1914219707 @default.
• W2263177918 cites W1964514974 @default.
• W2263177918 cites W1981511184 @default.
• W2263177918 cites W1986007546 @default.
• W2263177918 cites W1987091651 @default.
• W2263177918 cites W2005447957 @default.
• W2263177918 cites W2013502943 @default.
• W2263177918 cites W2017337590 @default.
• W2263177918 cites W2019263612 @default.
• W2263177918 cites W2031823405 @default.
• W2263177918 cites W2040884411 @default.
• W2263177918 cites W2041657594 @default.
• W2263177918 cites W2087347434 @default.
• W2263177918 cites W2097839764 @default.
• W2263177918 cites W2099862356 @default.
• W2263177918 cites W2102831150 @default.
• W2263177918 cites W2104472920 @default.
• W2263177918 cites W2113362355 @default.
• W2263177918 cites W2117866949 @default.
• W2263177918 cites W2119479037 @default.
• W2263177918 cites W2119821739 @default.
• W2263177918 cites W2121033924 @default.
• W2263177918 cites W2123068691 @default.
• W2263177918 cites W2127069950 @default.
• W2263177918 cites W2132820034 @default.
• W2263177918 cites W2137226992 @default.
• W2263177918 cites W2139338362 @default.
• W2263177918 cites W2140069289 @default.
• W2263177918 cites W2140095548 @default.
• W2263177918 cites W2142387771 @default.
• W2263177918 cites W2145295623 @default.
• W2263177918 cites W2148603752 @default.
• W2263177918 cites W2150772522 @default.
• W2263177918 cites W2153138111 @default.
• W2263177918 cites W2154630456 @default.
• W2263177918 cites W2157083019 @default.
• W2263177918 cites W2158001550 @default.
• W2263177918 cites W2160354932 @default.
• W2263177918 cites W2160392347 @default.
• W2263177918 cites W2161195767 @default.
• W2263177918 cites W2163302275 @default.
• W2263177918 cites W2164535072 @default.
• W2263177918 cites W2170356051 @default.
• W2263177918 cites W2579923771 @default.
• W2263177918 cites W2912008377 @default.
• W2263177918 cites W2914238450 @default.
• W2263177918 cites W2989661724 @default.
• W2263177918 cites W3023786531 @default.
• W2263177918 cites W3120740533 @default.
• W2263177918 cites W3142838532 @default.
• W2263177918 cites W45695613 @default.
• W2263177918 hasPublicationYear "2010" @default.
• W2263177918 type Work @default.
• W2263177918 sameAs 2263177918 @default.
• W2263177918 citedByCount "0" @default.
• W2263177918 crossrefType "journal-article" @default.
• W2263177918 hasAuthorship W2263177918A5058849006 @default.
• W2263177918 hasAuthorship W2263177918A5089765894 @default.
• W2263177918 hasConcept C112972136 @default.
• W2263177918 hasConcept C11413529 @default.
• W2263177918 hasConcept C114614502 @default.
• W2263177918 hasConcept C115903097 @default.
• W2263177918 hasConcept C119857082 @default.
• W2263177918 hasConcept C122280245 @default.
• W2263177918 hasConcept C12267149 @default.
• W2263177918 hasConcept C134517425 @default.
• W2263177918 hasConcept C140417398 @default.
• W2263177918 hasConcept C154945302 @default.
• W2263177918 hasConcept C24138899 @default.
• W2263177918 hasConcept C33923547 @default.
• W2263177918 hasConcept C39891107 @default.
• W2263177918 hasConcept C41008148 @default.
• W2263177918 hasConcept C73555534 @default.
• W2263177918 hasConcept C74193536 @default.
• W2263177918 hasConceptScore W2263177918C112972136 @default.
• W2263177918 hasConceptScore W2263177918C11413529 @default.
• W2263177918 hasConceptScore W2263177918C114614502 @default.
• W2263177918 hasConceptScore W2263177918C115903097 @default.
• W2263177918 hasConceptScore W2263177918C119857082 @default.
• W2263177918 hasConceptScore W2263177918C122280245 @default.
• W2263177918 hasConceptScore W2263177918C12267149 @default.

• next