{"id":732,"date":"2017-03-12T12:11:00","date_gmt":"2017-03-12T12:11:00","guid":{"rendered":"http:\/\/www.kelp-ml.org\/?page_id=732"},"modified":"2017-03-13T15:30:22","modified_gmt":"2017-03-13T15:30:22","slug":"kernel-compositions","status":"publish","type":"page","link":"http:\/\/www.kelp-ml.org\/?page_id=732","title":{"rendered":"Kernel Compositions"},"content":{"rendered":"

KernelComposition<\/a>s\u00a0operate enriching the\u00a0computation of another kernel. Details on these popular kernels can be found in (Shawe-Taylor and Cristianini, 2004).<\/p>\n

\n

Polynomial Kernel<\/h3>\n

Java class<\/strong>: PolynomialKernel<\/a><\/p>\n

Source code<\/strong>: PolynomialKernel.java<\/a><\/p>\n

Maven Project<\/strong>: kelp-core<\/a><\/p>\n

JSON type<\/strong>:\u00a0poly<\/p>\n

Description<\/strong>: it\u00a0implicitly works in a features space where all the polynomials of the original features are available. As an example, a \"2^{nd}\" degree polynomial kernel applied over a linear kernel on\u00a0vector representations will automatically consider pairs of features in its similarity evaluation. Given a base kernel K<\/em>, \u00a0the\u00a0PolynomialKernel<\/a> applies the following\u00a0formula:<\/p>\n

\"\(poly_K(x,y)=\big ( aK(x,y)+b \big ) ^d\) \"<\/p>\n

where\u00a0\"d, a, b\" are kernel parameters. Common values are\u00a0\"d=2\",\u00a0\"a=1\" and\u00a0\"b=0\".<\/p>\n

Parameters<\/strong>:<\/p>\n