#### Filter Results:

- Full text PDF available (17)

#### Publication Year

1999

2017

- This year (1)
- Last 5 years (1)
- Last 10 years (1)

#### Publication Type

#### Co-author

#### Journals and Conferences

Learn More

We introduce the notion of kernel-alignment, a measure of similarity between two kernel functions or between a kernel and a target function. This quantity captures the degree of agreement between aâ€¦ (More)

We present an algorithm based on convex optimization for constructing kernels for semi-supervised learning. The kernel matrices are derived from the spectral decomposition of graph Laplacians, andâ€¦ (More)

The perceptron algorithm with margins is a simple, fast and effective learning algorithm for linear classifiers; it produces decision hyperplanes within some constant ratio of the maximal margin. Inâ€¦ (More)

- Steve R. Gunn, Jaz S. Kandola
- Machine Learning
- 2002

A widely acknowledged drawback of many statistical modelling techniques, commonly used in machine learning, is that the resulting model is extremely difficult to interpret. A number of new conceptsâ€¦ (More)

- Koby Crammer, Jaz S. Kandola, Yoram Singer
- NIPS
- 2003

Online algorithms for classification often require vast amounts of memory and computation time when employed in conjunction with kernel functions. In this paper we describe and analyze a simpleâ€¦ (More)

- Jaz S. Kandola, John Shawe-Taylor, Nello Cristianini
- NIPS
- 2002

The standard representation of text documents as bags of words suffers from well known limitations, mostly due to its inability to exploit semantic similarity between terms. Attempts to incorporateâ€¦ (More)

- John Shawe-Taylor, Christopher K. I. Williams, Nello Cristianini, Jaz S. Kandola
- IEEE Transactions on Information Theory
- 2005

In this paper, the relationships between the eigenvalues of the m/spl times/m Gram matrix K for a kernel /spl kappa/(/spl middot/,/spl middot/) corresponding to a sample x/sub 1/,...,x/sub m/ drawnâ€¦ (More)

- Nello Cristianini, John Shawe-Taylor, Jaz S. Kandola
- NIPS
- 2001

In this paper we introduce new algorithms for unsupervised learning based on the use of a kernel matrix. All the information required by such algorithms is contained in the eigenvectors of the matrixâ€¦ (More)

- John Shawe-Taylor, Christopher K. I. Williams, Nello Cristianini, Jaz S. Kandola
- Discovery Science
- 2002

In this paper we analyze the relationships between the eigenvalues of the m Ã— m Gram matrix K for a kernel k(Â·, Â·) corresponding to a sample x1, . . . ,xm drawn from a density p(x) and theâ€¦ (More)

- John Shawe-Taylor, Nello Cristianini, Jaz S. Kandola
- NIPS
- 2001

We consider the problem of measuring the eigenvalues of a randomly drawn sample of points. We show that these values can be reliably estimated as can the sum of the tail of eigenvalues. Furthermore,â€¦ (More)