Shi, QinfengPetterson, JamesDror, GideonLangford, JohnSmola, AlexanderStrehl, AlexVishwanathan, S.V.N.2015-12-10April 16-1097273581Xhttp://hdl.handle.net/1885/57341We propose hashing to facilitate efficient kernels. This generalizes previous work using sampling and we show a principled way to compute the kernel matrix for data streams and sparse feature spaces. Moreover, we give deviation bounds from the exact kernel matrix. This has applications to estimation on strings and graphs.Keywords: Data stream; Feature space; Kernel matrices; Software engineering; Artificial intelligence20092016-02-24