Hash Kernels
نویسندگان
چکیده
We 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.
منابع مشابه
Hash Kernels for Structured Data
We 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.
متن کاملCirculant Preconditioners Constructed From Kernels
We consider circulant preconditioners for Hermitian Toeplitz systems from the view point of function theory. We show that some well-known circulant preconditioners can be derived from convoluting the generating function f of the Toeplitz matrix with famous kernels like the Dirichlet and the Fej er kernels. Several circulant precondition-ers are then constructed using this approach. Finally, we ...
متن کاملToeplitz Operators and Weighted Bergman Kernels
For a smoothly bounded strictly pseudoconvex domain, we describe the boundary singularity of weighted Bergman kernels with respect to weights behaving like a power (possibly fractional) of a defining function, and, more generally, of the reproducing kernels of Sobolev spaces of holomorphic functions of any real order. This generalizes the classical result of Fefferman for the unweighted Bergman...
متن کاملMin-Hash Fingerprints for Graph Kernels: A Trade-off among Accuracy, Efficiency, and Compression
Graph databases that emerge from several relevant scenarios (e.g., social networks, the Web) require powerful data management algorithms and techniques. A fundamental operation in graph data management is computing the similarity between two graphs. However, due to the large scale and high dimensionality of real graph databases, computing graph similarity becomes a challenging problem in real s...
متن کاملAn Improved Hash Function Based on the Tillich-Zémor Hash Function
Using the idea behind the Tillich-Zémor hash function, we propose a new hash function. Our hash function is parallelizable and its collision resistance is implied by a hardness assumption on a mathematical problem. Also, it is secure against the known attacks. It is the most secure variant of the Tillich-Zémor hash function until now.
متن کاملThe submission is for the special topic on Large Scale Learning Hash Kernels for Structured Data
We 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.
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2009