Abstract. In our previous work we have shown that Mahalanobis kernels are useful for support vector classifiers both from generalization ability and model selection speed. In this paper we propose using Mahalanobis kernels for function approximation. We determine the covariance matrix for the Mahalanobis kernel using all the training data. Model selection is done by line search. Namely, first the margin parameter and the error threshold are optimized and then the kernel parameter is optimized. According to the computer experiments for four benchmark problems, estimation performance of a Mahalanobis kernel with a diagonal covariance matrix optimized by line search is comparable to or better than that of an RBF kernel optimized by grid search.
Low-resistivity transparent indium tin oxide (ITO) films were fabricated on flexible polymer substrates by RF-magnetron sputtering. Hydrogen addition to the sputtering gas was effective in reducing the resistivity of ITO films deposited at room temperature. Resistivity was further decreased by inserting a SiO2 buffer layer between the substrate and ITO films. By optimizing the hydrogen pressure and SiO2 thickness, a resistivity of 3.4×10-4 Ω·cm was realized with a thickness of about 100 nm while maintaining an optical transparency of more than 85% in the visible range of the optical spectrum.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.