Maximum Margin Algorithms with Boolean Kernels Export

@misc{Khardon:2005, abstract = {Recent work has introduced Boolean kernels with which one can learn over a feature space containing all conjunctions of length up to k (for any 1 k n) over the original n Boolean features in the input space. This motivates the question of whether maximum margin algorithms such as support vector machines can learn Disjunctive Normal Form expressions in the PAC learning model using this kernel. We study this question, as well as a variant in which structural risk minimization (SRM) is performed where the class hierarchy is taken over the length of conjunctions.}, author = {Khardon, Roni and Servedio, Rocco A.}, citeulike-article-id = {5935633}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.731}, journal = {Journal of Machine Learning Research}, keywords = {boolean-kernels, kernel-trick, learning, phd, svm}, pages = {1405--1429}, posted-at = {2009-10-13 11:10:43}, priority = {5}, title = {Maximum Margin Algorithms with Boolean Kernels}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.731}, volume = {6}, year = {2005} }

TY - ELEC ID - Khardon:2005 L3 - citeulike-article-id:5935633 N2 - Recent work has introduced Boolean kernels with which one can learn over a feature space containing all conjunctions of length up to k (for any 1 k n) over the original n Boolean features in the input space. This motivates the question of whether maximum margin algorithms such as support vector machines can learn Disjunctive Normal Form expressions in the PAC learning model using this kernel. We study this question, as well as a variant in which structural risk minimization (SRM) is performed where the class hierarchy is taken over the length of conjunctions. JF - Journal of Machine Learning Research EP - 1429 TI - Maximum Margin Algorithms with Boolean Kernels VL - 6 SP - 1405 KW - boolean-kernels KW - kernel-trick KW - learning KW - phd KW - svm AU - Khardon, Roni AU - Servedio, Rocco PY - 2005/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.731 ER -