John Shawe-taylor

Stability Analysis of Kernel Canonical Correlation Analysis: Theory and Practice (2008)

Hardoon, David, Shawe-Taylor, John

Canonical Correlation Analysis is a technique for finding pairs of basis vectors that maximise the correlation of a set of paired variables, these pairs can be considered as two views of the same...

Kernel Regression Framework for Machine Translation: UCL System Description for WMT 2008 Shared Translation Task. (2008)

Wang, Zhuoran, Shawe-Taylor, John

The novel kernel regression model for SMT only demonstrated encouraging results on small-scale toy data sets in previous works due to the complexities of kernel methods. It is the first time results...

Handbook for the GREAT08 Challenge: An image analysis competition for cosmological lensing (2008)

Bridle, Sarah, Shawe-Taylor, John, Amara, Adam, Applegate, Douglas, Balan, Sreekumar T., Berge, Joel, ...

The GRavitational lEnsing Accuracy Testing 2008 (GREAT08) Challenge focuses on a problem that is of crucial importance for future observations in cosmology. The shapes of distant galaxies can be used...

Using String Kernels to Identify Famous Performers from their Playing Style (2008)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this paper we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Kernel-based machine translation (2008)

Wang, Zhuoran, Shawe-Taylor, John

In this chapter, we introduce a novel machine translation framework based on kernel regression techniques. In our model, the translation task is viewed as a string-to-string mapping, for which ridge...

Using String Kernels to Identify Famous Performers from their Playing Style (2008)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this paper we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Using String Kernels to Identify Famous Performers from their Playing Style (2008)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this paper we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Using String Kernels to Identify Famous Performers from their Playing Style (2008)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this paper we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Can style be learned? A machine learning approach towards ‘performing’ as famous pianists (2007)

Dorard, Louis, Hardoon, David, Shawe-Taylor, John

In this paper a novel method for performing music in the style of famous pianists is presented. We use Kernel Canonical Correlation Analysis (KCCA), a method which looks for a common semantic...

Variational Inference for Diffusion Processes (2007)

Archambeau, Cedric, Opper, Manfred, Shen, Yuan, Cornford, Dan, Shawe-Taylor, John

Diffusion processes are a family of continuous-time continuous-state stochastic processes that are in general only partially observed. The joint estimation of the forcing parameters and the system...

Sparse Canonical Correlation Analysis (2007)

Hardoon, David, Shawe-Taylor, John

In this paper we present a novel method for solving Canonical Correlation Analysis (CCA) in a sparse convex framework using a least squares approach. The presented method focuses on the scenario when...

Sparse CCA for Bilingual Word Generation (2007)

Hardoon, David, Shawe-Taylor, John

Proposed by Hotelling 1936, Canonical Correlation Analysis (CCA) is a technique for finding pairs of basis vectors that maximises the correlation between a set of paired variables. The set of paired...

Revised Loss Bounds for the Set Covering Machine and Sample-Compression Loss Bounds for Imbalanced Data (2007)

Hussain, Zakria, Laviolette, Francois, Marchand, Mario, Shawe-Taylor, John, Brubaker, Spencer Charles, Mullin, Matthew D

Marchand and Shawe-Taylor (2002) have proposed a loss bound for the set covering machine that has the property to depend on the observed fraction of positive examples and on what the classifier...

A Framework for Probability Density Estimation (2007)

Shawe-Taylor, John, Dolia, Alexander N.

The paper introduces a new framework for learning probability density functions. A theoretical analysis suggests that we can tailor a distribution for a class of tasks by training it to fit a small...

Evaluation of Variational and Markov Chain Monte Carlo Methods for Inference in Partially Observed Stochastic Dynamic Systems (2007)

Shen, Yuan, Archambeau, Cedric, Cornford, Dan, Opper, Manfred, Shawe-Taylor, John, Barillec, Remi

In recent work we have developed a novel variational inference method for partially observed systems governed by stochastic differential equations. In this paper we provide a comparison of the...

Using String Kernels to Identify Famous Performers from their Playing Style (2007)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this paper we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Exploiting the Prior in the PAC-Bayes Bound (2007)

Parrado-Hernandez, Emilio, Shawe-Taylor, John, Ambroladze, Amiran

This paper presents two SVM-like classification algorithms whose design criterion is to minimise the PAC-Bayes bound instead of to maximise the classification margin. A main goal of this work is to...

Using generalization error bounds to train the set covering machine (2007)

Hussain, Zakria, Shawe-Taylor, John

In this paper we eliminate the need for parameter estimation associated with the set covering machine (SCM) by directly minimizing generalization error bounds. Firstly, we consider a sub-optimal...

Bounding the k-family-wise error rate using resampling methods (2007)

De Bie, Tijl, Shawe-Taylor, John

The multiple hypothesis testing (MHT) problem has long been tackled by controlling the family-wise error rate (FWER), which is the probability that any of the hypotheses tested is unjustly rejected....

Using Image Stimuli to Drive fMRI Analysis (2007)

Hardoon, David, Mourao-Miranda, Janaina, Brammer, Michael, Shawe-Taylor, John

We introduce a new unsupervised fMRI analysis method based on Kernel Canonical Correlation Analysis which differs from the class of supervised learning methods that are increasingly being employed in...

Stability Analysis of Kernel Canonical Correlation Analysis: Theory and Practice (2007)

Hardoon, David, Shawe-Taylor, John

Canonical Correlation Analysis is a technique for finding pairs of basis vectors that maximise the correlation of a set of paired variables, these pairs can be considered as two views of the same...

Kernel regression based machine translation (2007)

Wang, Zhuoran, Shawe-Taylor, John, Szedmak, Sandor

We present a novel machine translation framework based on kernel regression techniques. In our model, the translation task is viewed as a string-to-string mapping, for which a regression type...

Kernel regression based machine translation (2007)

Wang, Zhuoran, Shawe-Taylor, John, Szedmak, Sandor

We present a novel machine translation framework based on kernel regression techniques. In our model, the translation task is viewed as a string-to-string mapping, for which a regression type...

A Framework for Probability Density Estimation (2007)

Shawe-Taylor, John, Dolia, Alexander

The paper introduces a new framework for learning probability density functions. A theoretical analysis suggests that we can tailor a distribution for a class of tasks by training it to fit a small...

Margin based Transductive Graph Cuts using Linear Programming (2007)

Pelckmans, Kristiaan, Shawe-Taylor, John, Suykens, Johan, De Moor, Bart

This paper studies the problem of inferring a partition (or a graph cut) of an undirected deterministic graph where the labels of some nodes are observed - thereby bridging a gap between graph theory...

Unsupervised analysis of fMRI data using Kernel Canonical Correlation (2007)

Hardoon, David, Mourao-Miranda, Janaina, Brammer, Michael, Shawe-Taylor, John

We introduce a new unsupervised fMRI analysis method based on Kernel Canonical Correlation Analysis which differs from the class of supervised learning methods ( e,g the Support Vector Machine) that...

NEW FEATURE SELECTION FRAMEWORKS IN EMOTION RECOGNITION TO EVALUATE THE INFORMATIVE POWER OF SPEECH RELATED FEATURES (2007)

Altun, Halis, Shawe-Taylor, John, Polat, Gokhan

In this paper, we propose two new frameworks, so as to boost the feature selection algorithms in a way that the selected features will be more informative in terms of class-separability. In the first...

Kernel Methods: A paradigm for Pattern Analysis (2007)

Saunders, Craig, Cristianini, Nello, Shawe-Taylor, John

An introductory chapter describing how the use of kernel methods has become a paradigm for pattern analysis in many different application areas.

Kernel methods (2007)

Cristianini, Nello, Shawe-Taylor, John, Saunders, Craig

This is a Kernel Methods overview/tutorial chapter.

Tighter PAC-Bayes Bounds (2007)

Ambroladze, Amiran, Parrado-Hernandez, Emilio, Shawe-Taylor, John

This paper proposes a PAC-Bayes bound to measure the performance of Support Vector Machine (SVM) classifiers. The bound is based on learning a prior over the distribution of classifiers with a part...

Kernel ellipsoidal trimming (2007)

Dolia, Alexander, Harris, Chris, Shawe-Taylor, John, Titterington, Mike

Ellipsoid estimation is important in many practical areas such as control, system identification, visual/audio tracking, experimental design, data mining, robust statistics and statistical outlier or...

Gaussian Process Approximations of Stochastic Differential Equations (2007)

Archambeau, Cedric, Cornford, Dan, Opper, Manfred, Shawe-Taylor, John

Stochastic differential equations arise naturally in a range of contexts, from financial to environmental modeling. Current solution methods are limited in their representation of the posterior...

Linear Programming Boosting for the Classification of Musical Genre (2007)

Diethe, Tom, Shawe-Taylor, John

area of music research, and as such provides a good starting point for testing a new algorithm. The Music Information Retrieval Evaluation eXchange (MIREX) is a yearly competition in a wide range of...

Gaussian Process Approximations of Stochastic Differential Equation (2007)

Archambeau, Cedric, Cornford, Dan, Opper, Manfred, Shawe-Taylor, John

Stochastic differential equations arise naturally in a range of contexts, from financial to environmental modeling. Current solution methods are limited in their representation of the posterior...

Variational Inference for Diffusion Processes (2007)

Archambeau, Cedric, Opper, Manfred, Shen, Yuan, Cornford, Dan, Shawe-Taylor, John

Diffusion processes are a family of continuous-time continuous-state stochastic processes that are in general only partly observed. The joint estimation of the forcing parameters and the system noise...

Efficient algorithms for max-margin structured classification (2007)

Rousu, Juho, Saunders, Craig, Szedmak, Sandor, Shawe-Taylor, John

We present a general and efficient optimisation methodology for for max-margin sructured classification tasks. The efficiency of the method relies on the interplay of several techiques:...

Information Retrieval by Inferring Implicit Queries from Eye Movements (2006)

Hardoon, David, Shawe-Taylor, John, Ajanki, Antti, Puolamäki, Kai, Kaski, Samuel

We introduce a new search strategy, in which the information retrieval (IR) query is inferred from eye movements measured when the user is reading text during an IR task. In training phase, we know...

Unsupervised fMRI Analysis (2006)

Hardoon, David, Mourao-Miranda, Janaina, Brammer, Michael, Shawe-Taylor, John

Recently machine learning methodology has been used increasing to analyze the relationship between stimulus categories and fMRI responses. Here, we introduce a new unsupervised machine learning...

A Framework for Probability Density Estimation (2006)

Shawe-Taylor, John, Dolia, Alexander N.

The paper introduces a new framework for learning probability density functions. A theoretical analysis suggests that we can tailor a distribution for a class of tasks by training it to fit a small...

The Minimum Volume Covering Ellipsoid Estimation in Kernel-Defined Feature Spaces (2006)

Dolia, Alexander, De Bie, Tijl, Harris, Chris, Shawe-Taylor, John, Titterington, Mike

Minimum volume covering ellipsoid estimation is important in areas such as systems identification, control, video tracking, sensor management, and novelty detection. It is well known that finding the...

Approximate Maximum Margin Algorithms with Rules Controlled by the Number of Mistakes (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a family of Perceptron-like algorithms with margin in which both the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, and the...

Approximate Maximum Margin Algorithms with Rules Controlled by the Number of Mistakes (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a family of Perceptron-like algorithms with margin in which both the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, and the...

Approximate Maximum Margin Algorithms with Rules Controlled by the Number of Mistakes (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a family of Perceptron-like algorithms with margin in which both the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, and the...

Sparse Feature Extraction using Generalised Partial Least Squares (2006)

Dhanjal, Charanpal, Gunn, Steve, Shawe-Taylor, John

We describe a general framework for feature extraction based on the deflation scheme used in Partial Least Squares (PLS). The framework provides many desirable properties, such as conjugacy and...

Approximate Maximum Margin Algorithms with Rules Controlled by the Number of Mistakes (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a family of Perceptron-like algorithms with margin in which both the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, and the...

Efficient algorithms for max-margin structured classification (2006)

Saunders, Craig, Szedmak, Sandor, Shawe-Taylor, John, Rousu, Juho

We present a general and efficient optimization methodology for max-margin structured classification tasks. The efficiency of the method relies on the interplay of several techniques: formulation of...

Gaussian Process Approximations of Stochastic Differential Equations (2006)

Archambeau, Cedric, Cornford, Dan, Opper, Manfred, Shawe-Taylor, John

Stochastic differential equations arise naturally in a range of contexts, from financial to environmental modelling. Current solution methods are based on a range of strong and weak approximation...

Using String Kernels to Identify Famous Performers from their Playing Style (2006)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this chapter we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Synthesis of Maximum Margin and Multiview Learning using Unlabeled Data (2006)

Szedmak, Sandor, Shawe-Taylor, John

In this presentation we show the semi-supervised learning with two input sources can be transformed into a maximum margin problem to be similar to a binary SVM. Our formulation exploits the unlabeled...

Learning via Linear Operators: Maximum Margin Regression (2006)

Szedmak, Sandor, Shawe-Taylor, John, Parado-Hernandez, Emilio

We introduce a maximum margin framework realizing a regression type learning in an arbitrary Hilbert space whilst the corresponding dual problem preserving the structure and, therefore, the...

Constant Rate Approximate Maximum Margin Algorithms (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a new class of perceptron-like algorithms with margin in which the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, remains...

Sparse Feature Extraction using Generalised Partial Least Squares (2006)

Dhanjal, Charanpal, Gunn, Steve R., Shawe-Taylor, John

We describe a general framework for feature extraction based on the deflation scheme used in Partial Least Squares (PLS). The framework provides many desirable properties, such as conjugacy and...

Constant Rate Approximate Maximum Margin Algorithms (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a new class of perceptron-like algorithms with margin in which the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, remains...

Sparse Feature Extraction using Generalised Partial Least Squares (2006)

Dhanjal, Charanpal, Gunn, Steve R., Shawe-Taylor, John

We describe a general framework for feature extraction based on the deflation scheme used in Partial Least Squares (PLS). The framework provides many desirable properties, such as conjugacy and...

Constant Rate Approximate Maximum Margin Algorithms (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a new class of perceptron-like algorithms with margin in which the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, remains...

The 2005 PASCAL Visual Object Classes Challenge (2006)

Everingham, Mark, Zisserman, Andrew, Williams, Christopher, Van Gool, Luc, Allan, Moray, Bishop, Chris, ...

The PASCAL Visual Object Classes Challenge ran from February to March 2005. The goal of the challenge was to recognize objects from a number of visual object classes in realistic scenes (i.e. not...

Sparse Feature Extraction using Generalised Partial Least Squares (2006)

Dhanjal, Charanpal, Gunn, Steve R., Shawe-Taylor, John

We describe a general framework for feature extraction based on the deflation scheme used in Partial Least Squares (PLS). The framework provides many desirable properties, such as conjugacy and...

Constant Rate Approximate Maximum Margin Algorithms (2006)

Tsampouka, Petroula, Shawe-Taylor, John

We present a new class of perceptron-like algorithms with margin in which the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, remains...

Sparse Feature Extraction using Generalised Partial Least Squares (2006)

Dhanjal, Charanpal, Gunn, Steve R., Shawe-Taylor, John

We describe a general framework for feature extraction based on the deflation scheme used in Partial Least Squares (PLS). The framework provides many desirable properties, such as conjugacy and...

Learning Hierarchies at Two-class Complexity (2005)

Szedmak, Sandor, Saunders, Craig, Shawe-Taylor, John, Rousu, Juho

It is assumed that to learn discriminative identification function when the output space is a labelled hierarchy is a much more complex problem than binary classification. In this presentation we...

Two view learning: SVM-2K, Theory and Practice (2005)

Farquhar, Jason, Hardoon, David, Meng, Hongying, Shawe-Taylor, John, Szedmak, Sandor

Kernel methods make it relatively easy to define complex high-dimensional feature spaces. This raises the question of how we can identify the relevant subspaces for a particular learning task. When...

Revised Loss Bound for the Set Covering Machine (2005)

Hussain, Zakria, Brubaker, Spencer Charles, Laviolette, Francois, Marchand, Mario, Shawe-Taylor, John

A revised loss bound for the Set Covering Machine in order to rectify the problem of imbalanced classifications.

Kernel-based Learning of Hierarchical Multilabel Classification Models (2005)

Rousu, Juho, Saunders, Craig, Szedmak, Sandor, Shawe-Taylor, John

We present a kernel-based algorithm for hierarchical text classification where the documents are allowed to belong to more than one category at a time. The classification model is a variant of the...

Using String Kernels to Identify Famous Performers from their Playing Style (2005)

Saunders, Craig, Hardoon, David, Shawe-Taylor, John, Widmer, Gerhard

In this paper we show a novel application of string kernels: that is to the problem of recognising famous pianists from their style of playing. The characteristics of performers playing the same...

Multiclass Learning at One-Class Complexity (2005)

Szedmak, Sandor, Shawe-Taylor, John

We show in this paper the multiclass classification problem can be implemented in the maximum margin framework with the complexity of one binary Support Vector Machine. We show reducing the...

The use of machine translation tools for cross-lingual text mining (2005)

Fortuna, Blaz, Shawe-Taylor, John

Eigen-analysis such as LSI or KCCA was already successfully applied to cross-lingual information retrieval. This approach has a weakness in that it needs an aligned training set of documents. In this...

Learning Hierarchical Multi-Category Text Classification Models (2005)

Rousu, Juho, Saunders, Craig, Szedmak, Sandor, Shawe-Taylor, John

We present a kernel-based algorithm for hierarchical text classification where the documents are allowed to belong to more than one category at a time. The classification model is a variant of the...

Prior Support Vector Machines: minimum-bound vs. maximum-margin classifiers (2005)

Ambroladze, Amiran, Parrado-Hernandez, Emilio, Shawe-Taylor, John

In this paper we introduce a new algorithm to train Support Vector Machines that aims at the minimisation of the PAC-Bayes bound on the error instead of at the traditional maximisation of the margin....

Learning the Prior for the PAC-Bayes Bound (2005)

Ambroladze, Amiran, Parrado-Hernandez, Emilio, Shawe-Taylor, John

This paper presents a bound on the performance of a Support Vector Machine obtained within the PAC-Bayes framework. The bound is computed by means of the estimation of a prior of the distribution of...

Complexity of Pattern Classes and Lipschitz Property (2005)

Ambroladze, Amiran, Parrado-Hernandez, Emilio, Shawe-Taylor, John

Rademacher and Gaussian complexities are successfully used in learning theory for measuring the capacity of the class of functions to be learnt. One of the most important properties for these...

An Investigation of Feature Models for Music Genre Classification using the Support Vector Classifier (2005)

Meng, Anders, Shawe-Taylor, John

In music genre classification the decision time is typically of the order of several seconds however most automatic music genre classification systems focus on short time features derived from...

Rademacher analysis of infimum classifiers (2005)

Ambroladze, Amiran, Parrado-Hernandez, Emilio, Shawe-Taylor, John

This paper addresses the problem of analysing the performance of classifiers obtained as the infimum of a set of k weak learners. The main result consists in a bound on the error of these classifiers...

Semantic Text Features from Small World Graphs (2005)

Leskovec, Jure, Shawe-Taylor, John

We present a set of methods for creating a semantic representation from a collection of textual documents. Given a document collection we use a simple algorithm to connect the documents into a tree...

Generic object recognition by combining distinct features in machine learning (2005)

Meng, Hongying, Hardoon, David, Shawe-Taylor, John, Szedmak, Sandor

In a generic image object recognition or categorization system, the relevant features or descriptors from a characteristic point, patch or region of an image are often obtained by different...