Quadrianto, Novi ; JFA; CORA, Smola, Alexander J., Song, Le, Tuytelaars, Tinne; U0018289 ;
Quadrianto N., Smola A.J., Song L., Tuytelaars T., ''Kernelized sorting'', IEEE transactions on pattern analysis and machine intelligence, 2010 (in press).
Robust Near-Isometric Matching via Structured Learning of Graphical Models (2009)
Julian J. Mcauley, Tibério S. Caetano, Alexander J. Smola
Models for near-rigid shape matching are typically based on distance-related features, in order to infer matches that are consistent with the isometric assumption. However, real shapes from image...
Robust Near-Isometric Matching via Structured Learning of Graphical Models (2008)
McAuley, Julian J., Caetano, Tiberio S., Smola, Alexander J.
Models for near-rigid shape matching are typically based on distance-related features, in order to infer matches that are consistent with the isometric assumption. However, real shapes from image...
Predicting Structured Data edited by (2008)
David S. Touretzky, John E. Moody, David S. Touretzky, Gökhan Bakır, Thomas Hofmann, Bernhard Schölkopf, ...
All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval)
Jiayuan Huang, Karsten M. Borgwardt, Alexander J. Smola, Arthur Gretton, Bernhard Schölkopf
We consider the scenario where training and test data are drawn from different distributions, commonly referred to as sample selection bias. Most algorithms for this setting try to first recover...
A Kernel Method for the Two-Sample Problem (2008)
Gretton, Arthur, Borgwardt, Karsten, Rasch, Malte J., Scholkopf, Bernhard, Smola, Alexander J.
We propose a framework for analyzing and comparing distributions, allowing us to design statistical tests to determine if two samples are drawn from different distributions. Our test statistic is the...
Choon Hui Teo, Sam Roweis, Amir Globerson, Alexander J. Smola
Incorporating invariances into a learning algorithm is a common problem in machine learning. We provide a convex formulation which can deal with arbitrary loss functions and arbitrary losses. In...
Jiayuan Huang, Karsten M. Borgwardt, Alexander J. Smola, Arthur Gretton, Bernhard Schölkopf
We consider the scenario where training and test data are drawn from different distributions, commonly referred to as sample selection bias. Most algorithms for this setting try to first recover...
Predicting Structured Data edited by (2008)
David S. Touretzky, John E. Moody, David S. Touretzky, Gökhan Bakır, Thomas Hofmann, Bernhard Schölkopf, ...
All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval)
Arthur Gretton, Bernhard Schölkopf, Karsten M. Borgwardt, Malte Rasch, Alexander J. Smola
We propose two statistical tests to determine if two samples are from different distributions. Our test statistic is in both cases the distance between the means of the two samples mapped into a...
Jiayuan Huang, Karsten M. Borgwardt, Alexander J. Smola, Arthur Gretton, Bernhard Schölkopf
We consider the scenario where training and test data are drawn from different distributions, commonly referred to as sample selection bias. Most algorithms for this setting try to first recover...
and Australian National University (2008)
Cheng Soon Ong, Alexander J. Smola, Robert C. Williamson, Ralf Herbrich
This paper addresses the problem of choosing a kernel suitable for estimation with a support vector machine, hence further automating machine learning. This goal is achieved by defining a reproducing...
Arthur Gretton, Le Song, Kenji Fukumizu, Bernhard Schölkopf, Choon Hui Teo, Alexander J. Smola
Although kernel measures of independence have been widely applied in machine learning (notably in kernel ICA), there is as yet no method to determine whether they have detected statistically...
S. V. N. Vishwanathan, Alexander J. Smola
In this paper we present a new algorithm suitable for matching discrete objects such as strings and trees in linear time, thus obviating dynamic programming with quadratic time complexity....
Learning the Kernel with Hyperkernels Cheng Soon Ong (2008)
Alexander J. Smola, Robert C. Williamson
Editor: U.N. Known (joint publication with www.kernel-machines.org) This paper addresses the problem of choosing a kernel suitable for estimation with a Support Vector Machine, hence further...
Arthur Gretton, Bernhard Schölkopf, Karsten M. Borgwardt, Malte Rasch, Alexander J. Smola
We propose two statistical tests to determine if two samples are from different distributions. Our test statistic is in both cases the distance between the means of the two samples mapped into a...
Arthur Gretton, Bernhard Schölkopf, Karsten M. Borgwardt, Malte Rasch, Alexander J. Smola
We propose two statistical tests to determine if two samples are from different distributions. Our test statistic is in both cases the distance between the means of the two samples mapped into a...
1.2 A Simple Pattern Recognition Algorithm............... 3 (2008)
Bernhard Schölkopf, Alexander J. Smola, London England, Alexander J. Smola
All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without...
Predicting Structured Data edited by (2008)
David S. Touretzky, John E. Moody, David S. Touretzky, Gökhan Bakır, Thomas Hofmann, Bernhard Schölkopf, ...
All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval)
S. V. N. Vishwanathan, Alexander J. Smola, M. Narasimha Murty
We present a fast iterative support vector training algorithm for a large variety of different formulations. It works by incrementally changing a candidate support vector set using a greedy approach,...
Jiayuan Huang, Karsten M. Borgwardt, Alexander J. Smola, Arthur Gretton, Bernhard Schölkopf
We consider the scenario where training and test data are drawn from different distributions, commonly referred to as sample selection bias. Most algorithms for this setting try to first recover...
Jiayuan Huang, Karsten M. Borgwardt, Alexander J. Smola, Arthur Gretton, Bernhard Schölkopf
We consider the scenario where training and test data are drawn from different distributions, commonly referred to as sample selection bias. Most algorithms for this setting try to first recover...
Choon Hui Teo, Sam Roweis, Amir Globerson, Alexander J. Smola
Incorporating invariances into a learning algorithm is a common problem in machine learning. We provide a convex formulation which can deal with arbitrary loss functions and arbitrary losses. In...
THE KERNEL MUTUAL INFORMATION Arthur Gretton MPI for Biological CyberneticsSpemannstr 38 (2008)
ABSTRACT We introduce a new contrast function, the kernel mutual informa-tion (KMI), to measure the degree of independence of continuous
A kernel statistical test of independence (2008)
Arthur Gretton, Kenji Fukumizu, Choon Hui Teo, Le Song, Bernhard Schölkopf, Alexander J. Smola
Although kernel measures of independence have been widely applied in machine learning (notably in kernel ICA), there is as yet no method to determine whether they have detected statistically...
Glenn M. Fung, Olvi L. Mangasarian, Alexander J. Smola
A finite concave minimization algorithm is proposed for constructing kernel classifiers that use a minimal number of data points both in generating and characterizing a classifier. The algorithm is...
Support Vector Machines and Kernel Algorithms (2007)
Alexander J. Smola, Alex Smola, Bernhard Schölkopf, Bernhard Schölkopf, Er J. Smola
One of the fundamental problems of learning theory is the following: suppose we are given two classes of objects. We are then faced with a new object, and we have to assign it to one of the two...
Alexander J. Smola, Risi Kondor
Abstract. We introduce a family of kernels on graphs based on the notion of regularization operators. This generalizes in a natural way the notion of regularization and Greens functions, as commonly...
Machine Learning using Hyperkernels (2007)
Cheng Soon Ong, Alexander J. Smola
We expand on the problem of learning a kernel via a RKHS on the space of kernels itself.
Cheng Soon Ong, Alexander J. Smola, Er J. Smola, Robert C. Williamson
We consider the problem of choosing a kernel suitable for estimation using a Gaussian Process estimator or a Support Vector Machine. A novel solution is presented which involves defining a...
Kernel methods in machine learning (2007)
Hofmann, Thomas, Schölkopf, Bernhard, Smola, Alexander J.
We review machine learning methods employing positive definite kernels. These methods formulate learning and estimation problems in a reproducing kernel Hilbert space (RKHS) of functions defined on...
Correcting sample selection bias by unlabeled data (2007)
Jiayuan Huang, Arthur Gretton, Bernhard Schölkopf, Alexander J. Smola, Karsten M. Borgwardt
Second Order Cone Programming Approaches for Handling Missing and Uncertain Data (2006)
Shivaswamy, Pannagadatta K, Bhattacharyya, Chiranjib, Smola, Alexander J
We propose a novel second order cone programming formulation for designing robust classifiers which can handle uncertainty in observations. Similar formulations are also derived for designing...
Second Order Cone Programming Approaches for Handling Missing and Uncertain Data (2006)
Shivaswamy, Pannagadatta K, Bhattacharyya, Chiranjib, Smola, Alexander J
We propose a novel second order cone programming formulation for designing robust classifiers which can handle uncertainty in observations. Similar formulations are also derived for designing...
A Review of Kernel Methods in Machine Learning (2006)
Thomas Hofmann, Bernhard Schölkopf, Alexander J. Smola, Thomas Hofmann, Bernhard Schölkopf, Er J. Smola
Abstract. We review recent methods for learning with positive definite kernels. All these methods formulate learning and estimation problems as linear tasks in a reproducing kernel Hilbert space...
A Review of Kernel Methods in Machine Learning (2006)
Thomas Hofmann, Bernhard Schölkopf, Alexander J. Smola, Thomas Hofmann, Bernhard Schölkopf, Er J. Smola
Abstract. We review recent methods for learning with positive definite kernels. All these methods formulate learning and estimation problems as linear tasks in a reproducing kernel Hilbert space...
For Handling Missing, Pannagadatta K. Shivaswamy, Chiranjib Bhattacharyya, Alexander J. Smola, P. Bennett, Emilio Parrado-hernández
We propose a novel second order cone programming formulation for designing robust classifiers which can handle uncertainty in observations. Similar formulations are also derived for designing...
Nonparametric quantile estimation (2006)
Ichiro Takeuchi, Quoc V. Le, Timothy D. Sears, Alexander J. Smola, Chris Williams
In regression, the desired estimate of y|x is not always given by a conditional mean, although this is most common. Sometimes one wants to obtain a good estimate that satisfies the property that a...
A review of RKHS methods in machine learning (2006)
Thomas Hofmann, Bernhard Schölkopf, Alexander J. Smola
Over the last ten years, estimation and learning methods utilizing positive definite kernels have become rather popular, particularly in machine learning. Since these methods have a stronger...
S. V. N. Vishwanathan, René Vidal, Alexander J. Smola
Abstract. We derive a family of kernels on dynamical systems by applying the Binet-Cauchy theorem to trajectories of states. Our derivation provides a unifying framework for all kernels on dynamical...
Statistical Machine Learning Program, Canberra (2005)
Quoc V. Le, Tim Sears, Alexander J. Smola
In regression, the desired estimate of y|x is not always given by a conditional mean, although this is most common. Sometimes one wants to obtain a good estimate that satisfies the property that a...
S. V. N. Vishwanathan, Alexander J. Smola, René Vidal
Abstract. We propose a family of kernels based on the Binet-Cauchy theorem, and its extension to Fredholm operators. Our derivation provides a unifying framework for all kernels on dynamical systems...
Learning the kernel with hyperkernels (2005)
Cheng Soon Ong, Alexander J. Smola, Robert C. Williamson, Ralf Herbrich
This paper addresses the problem of choosing a kernel suitable for estimation with a Support Vector Machine, hence further automating machine learning. This goal is achieved by defining a Reproducing...
Statistical Machine Learning Program, Canberra (2005)
Quoc V. Le, Tim Sears, Alexander J. Smola
In regression, the desired estimate of y|x is not always given by a conditional mean, although this is most common. Sometimes one wants to obtain a good estimate that satisfies the property that a...
S. V. N. Vishwanathan, Alexander J. Smola, René Vidal
Abstract. We propose a family of kernels based on the Binet-Cauchy theorem, and its extension to Fredholm operators. Our derivation provides a unifying framework for all kernels on dynamical systems...
Gaussian process classification for segmenting and annotating sequences (2004)
Yasemin Altun, Thomas Hofmann, Alexander J. Smola
Many real-world classification tasks involve the prediction of multiple, inter-dependent class labels. A prototypical case of this sort deals with prediction of a sequence of labels for a sequence of...
Learning with Non-Positive Kernels (2004)
Cheng Soon Ong, Stéphane Canu, Alexander J. Smola
In this paper we show that many kernel methods can be adapted to deal with indefinite kernels, that is, kernels which are not positive semidefinite. They do not satisfy Mercer 's condition and...
Gaussian process classification for segmenting and annotating sequences (2004)
Yasemin Altun, Thomas Hofmann, Alexander J. Smola
Many real-world classification tasks involve the prediction of multiple, inter-dependent class labels. A prototypical case of this sort deals with prediction of a sequence of labels for a sequence of...
Online Learning with Kernels (2003)
Jyrki Kivinen, Alexander J. Smola, Robert C. Williamson
Kernel based algorithms such as support vector machines have achieved considerable success in various problems in the batch setting where all of the training data is available in advance. Support...
Bayesian kernel methods (2003)
Alexander J. Smola, Bernhard Schölkopf
Bayesian methods allow for a simple and intuitive representation of the function spaces used by kernel methods. This chapter describes the basic principles of Gaussian Processes, their implementation...
Fast Kernels for String and Tree Matching (2003)
Vishwanathan Dept Of, S. V. N. Vishwanathan, Alexander J. Smola
In this paper we present a new algorithm suitable for matching discrete objects such as strings and trees in linear time, thus obviating dynamic programming with quadratic time complexity....
Learning the Kernel with Hyperkernels (2003)
Cheng Soon Ong, Alexander J. Smola, Robert C. Williamson
This paper addresses the problem of choosing a kernel suitable for estimation with a Support Vector Machine, hence further automating machine learning. This goal is achieved by defining a Reproducing...
Logic, Trees and Kernels (2003)
Adam Kowalczyk, Alexander J. Smola, Robert C. Williamson
Kernel based methods achieved much of their initial success on problems with real valued attributes. There are many problems with discrete attributes (including Boolean) and in this paper we present...
Minimal kernel classifiers (2002)
Glenn M. Fung, Olvi L. Mangasarian, Alexander J. Smola
A finite concave minimization algorithm is proposed for constructing kernel classifiers that use a minimal number of data points both in generating and characterizing a classifier. The algorithm is...
Minimal kernel classifiers (2002)
Glenn M. Fung, Olvi L. Mangasarian, Alexander J. Smola
A finite concave minimization algorithm is proposed for constructing kernel classifiers that use a minimal number of data points both in generating and characterizing a classifier. The algorithm is...
Minimal kernel classifiers (2002)
Glenn M. Fung, Olvi L. Mangasarian, Alexander J. Smola
A finite concave minimization algorithm is proposed for constructing kernel classifiers that use a minimal number of data points both in generating and characterizing a classifier. The algorithm is...
Regularized Principal Manifolds (2001)
Alexander J. Smola, Sebastian Mika, Bernhard Schölkopf, Robert C. Williamson
Many settings of unsupervised learning can be viewed as quantization problems - the minimization of the expected quantization error subject to some restrictions. This allows the use of tools such as...
Advances in Large Margin Classifiers (2000)
Alexander J. Smola, Alex J. Smola, Peter Bartlett, Dale Schuurmans (Eds.), Peter Bartlett, Bernhard Schölkopf, ...
Contents Preface vii 1 Introduction to Large Margin Classifiers 1 Alex J. Smola, Peter Bartlett, Bernhard Scholkopf, and Dale Schuurmans 2 Large Margin Rank Boundaries for Ordinal Regression 29 Ralf...
Advances in Large Margin Classifiers (2000)
Alexander J. Smola, Alex J. Smola, Peter Bartlett, Peter Bartlett, Bernhard Scholkopf, Bernhard Scholkopf, ...
this article also provide a website to obtain the data
Advances in Large Margin Classifiers (2000)
Alexander J. Smola, Alex J. Smola, Peter Bartlett, Peter Bartlett, Bernhard Scholkopf, Bernhard Scholkopf, ...
this article also provide a website to obtain the data
Regularized principal manifolds (1999)
Alexander J. Smola, Robert C. Williamson
Editor: Douglas H. Fisher (joint publication with www.kernel-machines.org) Many settings of unsupervised learning can be viewed as quantization problems- the minimization of the expected quantization...
Generalization Bounds via Eigenvalues of the Gram Matrix (1999)
Bernhard Schölkopf, John Shawe-Taylor, Alexander J. Smola, Bernhard Scholkopf Gmd, Er J. Smola, Robert C. Williamson
Model selection in Support Vector machines is usually carried out by minimizing the quotient of the radius of the smallest enclosing sphere of the data and the observed margin on the training set. We...
Advances in Large Margin Classifiers (1999)
Alexander J. Smola, Rudower Chaussee
Introduction to Large Margin Classiers Alexander J. Smola GMD FIRST Rudower Chaussee 5 12489 Berlin, Germany smola@rst.gmd.de http://www.rst.gmd.de/smola Peter Bartlett Australian National...
Advances in Large Margin Classifiers (1999)
Alexander J. Smola, Alex J. Smola, Peter Bartlett, Dale Schuurmans (Eds.), Peter Bartlett, Bernhard Schölkopf, ...
this paper are taken from (Herbrich et al., 1999) Smola, Bartlett, Scholkopf, and Schuurmans: Advances in Large Margin Classifiers 1999/03/31 11:08
Advances in Kernel Methods - Support Vector Learning (1998)
Bernhard Schölkopf, Christopher J.C. Burges, Alexander J. Smola (Eds.), Alexander J. Smola
this paper should not be used as an indication of the quality of the method. The primary weakness of the MPM approaches is that they have not been guided by statistical learning theory. In the...
Prior Knowledge in Support Vector Kernels (1998)
Bernhard Schölkopf, Bernhard Sch Olkopf, Biologische Kybernetik, Patrice Simard, Vladimir Vapnik, Alexander J. Smola
We explore methods for incorporating prior knowledge about a problem at hand in Support Vector learning machines. We show that both invariances under group transformations and prior knowledge about...
From regularization operators to support vector kernels (1998)
Alexander J. Smola, Bernhard Schölkopf
We derive the correspondence between regularization operators used in Regularization Networks and Hilbert Schmidt Kernels appearing in Support Vector Machines. More specifically, we prove that the...
Chiranjib Bhattacharyya, Pannagadatta K. S, Alexander J. Smola
We propose a convex optimization based strategy to deal with uncertainty in the observations of a classification problem. We assume that instead of a sample (xi, yi) a distribution over (xi, yi) is...
Chiranjib Bhattacharyya, Pannagadatta K. S, Alexander J. Smola
We propose a mathematical programming method to deal with uncertainty in the observations of a classification problem. This means that we can deal with situations where instead of a sample (xi, yi)...