@string{ESANN = "Proceedings of the European symposium on artifical neural networks"} @string{ESANNadd = "Bruges, Belgium"} @inproceedings{Cawley2002a, author = "Cawley, G. C. and Talbot, N. L. C.", title = "Efficient formation of a basis in a kernel induced feature space", booktitle = ESANN # "~(ESANN 2002)", pages = "1--6", address = ESANNadd, month = apr, year = 2002, abstract = "Baudat and Anouar (2001) propose a simple greedy algorithm for estimation of an approximate basis of the subspace spanned by a set of fixed vectors embedded in a kernel induced feature space. The resulting set of basis vectors can then be used to construct sparse kernel expansions for classification and regression tasks. In this paper we describe five algorithmic improvements to the method of Baudat and Anouar, allowing the construction of an approximate basis with a computational complexity that is independent of the number of training patterns, depending only on the number of basis vectors extracted.", file = "esann2002a.pdf" } @article{Cawley2002b, author = "Cawley, G. C. and Talbot, N. L. C.", title = "Improved sparse least-squares support vector machines", journal = "Neurocomputing", volume = 48, pages = "1025--1031", month = oct, year = 2002, abstract = "Suykens \emph{et al.} describe a weighted least-squares formulation of the support vector machine for regression problems and presents a simple algorithm for sparse approximation of the typically fully dense kernel expansions obtained using this method. In this paper, we present an improved method for achieving sparsity in least-squares support vector machines, which takes into account the residuals for all training patterns, rather than only those incorporated in the sparse kernel expansion. The superiority of this algorithm is demonstrated on the motorcycle and Boston housing datasets.", file = "neurocomputing.pdf" } @inproceedings{Foxall2002, author = "Foxall, R. J. and Cawley, G. C. and Talbot, N. L. C. and Dorling, S. R. and Mandic, D. P.", title = "Heteroscedastic regularised kernel regression for prediction of episodes of poor air quality", booktitle = ESANN # "~(ESANN-2002)", pages = "19--24", address = ESANNadd, month = apr, year = 2002, abstract = "A regularised kernel regression model is introduced for data characterised by a heteroscedastic (input dependent variance) Gaussian noise process. The proposed model provides more robust estimates of the conditional mean than standard models and also confidence intervals (error bars) on predictions. The benefits of the proposed model are demonstrated for the task of non-linear prediction of episodes of poor air quality in urban environments.", file = "esann2002b.pdf" }and suppose you have a file

\documentclass[a4paper]{article} \usepackage{html} \begin{document} \nocite{*} \bibliographystyle{abshtml} \bibliography{sample} \end{document}Then the HTML page can be created using the

latex2html -split 0 -noinfo -nonavigation sampleIt's not perfect, but it saves typing it all directly into HTML.

Dr Nicola Talbot | School of Computing Sciences | University of East Anglia