Kernel Matching Pursuit for Large Datasets. Popovici, V., Bengio, S., & Thiran, J. Pattern Recognition, 38(12):2385–2390, 2005. ![Kernel Matching Pursuit for Large Datasets [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex Kernel Matching Pursuit is a greedy algorithm for building an approximation of a discriminant function as a linear combination of some basis functions selected from a kernel-induced dictionary. Here we propose a modification of the Kernel Matching Pursuit algorithm that aim s at making the method practical for large datasets. Starting from an approximating algorithm, the Weak Greedy Algorithm, we introduce a stochastic method for reducing the search space at each iteration. Then we study the implications of using an approximate algorithm and we show how one can control the trade-off between the accuracy and the need for resources. Finally we present some experiments performed on a large dataset that support our approach and illustrate its applicability.
@article{popovici:2005:pr,
author = {V. Popovici and S. Bengio and J.-P. Thiran},
title = {Kernel Matching Pursuit for Large Datasets},
journal = {Pattern Recognition},
volume = 38,
number = 12,
pages = {2385--2390},
year = 2005,
url = {publications/ps/popovici_2005_pr.ps.gz},
pdf = {publications/pdf/popovici_2005_pr.pdf},
djvu = {publications/djvu/popovici_2005_pr.djvu},
original= {2005/matching_pursuit_pr},
topics = {kernel},
web = {http://dx.doi.org/10.1016/j.patcog.2005.01.021},
abstract = {Kernel Matching Pursuit is a greedy algorithm for building an approximation of a discriminant function as a linear combination of some basis functions selected from a kernel-induced dictionary. Here we propose a modification of the Kernel Matching Pursuit algorithm that aim s at making the method practical for large datasets. Starting from an approximating algorithm, the Weak Greedy Algorithm, we introduce a stochastic method for reducing the search space at each iteration. Then we study the implications of using an approximate algorithm and we show how one can control the trade-off between the accuracy and the need for resources. Finally we present some experiments performed on a large dataset that support our approach and illustrate its applicability.},
categorie = {A},
}
Downloads: 0
{"_id":"kLJQLzeCe8RMxoxkq","bibbaseid":"popovici-bengio-thiran-kernelmatchingpursuitforlargedatasets-2005","downloads":0,"creationDate":"2018-01-22T16:01:07.603Z","title":"Kernel Matching Pursuit for Large Datasets","author_short":["Popovici, V.","Bengio, S.","Thiran, J."],"year":2005,"bibtype":"article","biburl":"http://bengio.abracadoudou.com/samy.bib","bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["V."],"propositions":[],"lastnames":["Popovici"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Bengio"],"suffixes":[]},{"firstnames":["J.-P."],"propositions":[],"lastnames":["Thiran"],"suffixes":[]}],"title":"Kernel Matching Pursuit for Large Datasets","journal":"Pattern Recognition","volume":"38","number":"12","pages":"2385–2390","year":"2005","url":"publications/ps/popovici_2005_pr.ps.gz","pdf":"publications/pdf/popovici_2005_pr.pdf","djvu":"publications/djvu/popovici_2005_pr.djvu","original":"2005/matching_pursuit_pr","topics":"kernel","web":"http://dx.doi.org/10.1016/j.patcog.2005.01.021","abstract":"Kernel Matching Pursuit is a greedy algorithm for building an approximation of a discriminant function as a linear combination of some basis functions selected from a kernel-induced dictionary. Here we propose a modification of the Kernel Matching Pursuit algorithm that aim s at making the method practical for large datasets. Starting from an approximating algorithm, the Weak Greedy Algorithm, we introduce a stochastic method for reducing the search space at each iteration. Then we study the implications of using an approximate algorithm and we show how one can control the trade-off between the accuracy and the need for resources. Finally we present some experiments performed on a large dataset that support our approach and illustrate its applicability.","categorie":"A","bibtex":"@article{popovici:2005:pr,\n author = {V. Popovici and S. Bengio and J.-P. Thiran},\n title = {Kernel Matching Pursuit for Large Datasets},\n journal = {Pattern Recognition},\n volume = 38,\n number = 12,\n pages = {2385--2390},\n year = 2005,\n url = {publications/ps/popovici_2005_pr.ps.gz},\n pdf = {publications/pdf/popovici_2005_pr.pdf},\n djvu = {publications/djvu/popovici_2005_pr.djvu},\n original= {2005/matching_pursuit_pr},\n topics = {kernel},\n web = {http://dx.doi.org/10.1016/j.patcog.2005.01.021},\n abstract = {Kernel Matching Pursuit is a greedy algorithm for building an approximation of a discriminant function as a linear combination of some basis functions selected from a kernel-induced dictionary. Here we propose a modification of the Kernel Matching Pursuit algorithm that aim s at making the method practical for large datasets. Starting from an approximating algorithm, the Weak Greedy Algorithm, we introduce a stochastic method for reducing the search space at each iteration. Then we study the implications of using an approximate algorithm and we show how one can control the trade-off between the accuracy and the need for resources. Finally we present some experiments performed on a large dataset that support our approach and illustrate its applicability.},\n categorie = {A},\n}\n\n","author_short":["Popovici, V.","Bengio, S.","Thiran, J."],"key":"popovici:2005:pr","id":"popovici:2005:pr","bibbaseid":"popovici-bengio-thiran-kernelmatchingpursuitforlargedatasets-2005","role":"author","urls":{"Paper":"http://bengio.abracadoudou.com/publications/ps/popovici_2005_pr.ps.gz"},"downloads":0},"search_terms":["kernel","matching","pursuit","large","datasets","popovici","bengio","thiran"],"keywords":[],"authorIDs":[],"dataSources":["9NCW2CDr4M3s5DvNX"]}