Kernel Machines

Prediction with Gaussian Processes: From Linear Regression to Linear Prediction and beyond

admin — 2008-05-13T08:48:55Z

Prediction with Gaussian Processes: From Linear Regression to Linear Prediction and beyond

admin — 2008-05-13T08:50:31Z

Financial Time Series Prediction using Least Squares Support Vector Machines within the Evidence Framework

admin — 2007-01-31T10:07:28Z

Predicting Time Series with Support Vector Machines

admin — 2007-01-31T10:07:30Z

SV regression with epsilon-insensitive and Huber loss functions. Experimental results on time-series prediction (Mackey-Glass and Santa Fe competition data set D, with a new record on the latter).

Prediction of Generalization Ability in Learning Machines

admin — 2007-01-31T10:07:37Z

In this thesis the soft margin classifier is introduced and discussed in great detail. Moreover the concept of effective VC-dimensions is introduced.

Nonlinear Prediction of Chaotic Time Series using a Support Vector Machine

admin — 2007-01-31T10:07:42Z

Uses SV regression on chaotic time-series (Mackey-Glass, Ikewda Map and Lorenz) and compares with other techniques reported (Casdalgi).

Prediction with Gaussian Processes: From Linear Regression to Linear Prediction and beyond

admin — 2007-01-31T10:09:02Z

Kernel Principal Component Regression with EM Approach to Nonlinear Principal Components Extraction.

admin — 2007-01-31T10:09:05Z

In kernel based methods such as Support Vector Machines, Kernel PCA, Gaussian Processes or Regularization Networks the computational requirements scale as O(n^3) where n is the number of training points. In this paper we investigate Kernel Principal Component Regression (KPCR) with the Expectation Maximization approach in estimating of the subset of p principal components (p < n) in a feature space defined by a positive definite kernel function. The computational requirements of the method are O(pn^2). Moreover, the algorithm can be implemented with memory requirements O(p^2)+O((p+1)n)). We give the theoretical description explaining how by the proper selection of a subset of non-linear principal components desired generalization of the KPCR is achieved. On two data sets we experimentally demonstrate this fact. Moreover, on a noisy chaotic Mackey-Glass time series prediction the best performance is achieved with p << n and experiments also suggests that in such cases we can also use significantly reduced training data sets to estimate the non-linear principal components. The theoretical relation and experimental comparison to Kernel Ridge Regression and epsilon-insensitive Support Vector Regression is also given.

The Generalized Bayesian Committee Machine

admin — 2007-01-31T10:09:09Z

In this paper we introduce the Generalized Bayesian Committee Machine (GBCM) for applications with large data sets. In particular, the GBCM can be used in the context of kernel based systems such as smoothing splines, kriging, regularization networks and Gaussian process regression which —for computational reasons— are otherwise limited to rather small data sets. The GBCM provides a novel and principled way of combining estimators trained for regression, classification, the prediction of counts, the prediction of lifetimes and other applications which can be derived from the exponential family of distributions. We describe an online version of the GBCM which only requires one pass through the data set and only requires the storage of a matrix of the dimension of the number of query or test points. After training, the prediction at additional test points only requires resources dependent on the number of query points but is independent of the number of training data. We confirm the good scaling behavior using real and experimental data sets.

A Novel Method of Protein Secondary Structure Prediction with High Segment Overlap Measure: Support Vector Machine Approach

admin — 2007-01-31T10:09:16Z

We introduced a new method of protein secondary structure prediction which is based on the theory of Support Vector Machine (SVM). SVM represents a new approach to supervised pattern classification which has been successfully applied to a wide range of pattern recognition problems, including object recognition, speaker identification, gene function prediction with microarray expression profile, etc. In these cases, the performance of SVM either matches or is significantly better than of traditional machine learning approaches, including neural networks. The first use of the SVM approach to predict protein secondary structure is described in this paper. Unlike the previous studies, we first constructed several binary classifiers, then assembled a tertiary classifier for three secondary structure states (helix, sheet and coil) based on these binary classifiers. The SVM method achieved a good performance of Segment Overlap accuracy SOV=76.2% through 7-fold cross validation on a database of 513 non-homologous protein chains, with multiple sequence alignments, which out-performs existing methods. Meanwhile three-state overall per-residue accuracy Q3 achieved 73.5%, which is at least comparable to the existing single prediction method. Furthermore a useful ¡®Reliability Index¡¯ for the predictions was developed. In addition, SVM has many attractive features, including effective avoidance of overfitting, the ability to handle large feature spaces, information condensing of the given data set, etc. The SVM method is conveniently applied to many other pattern classification tasks in biology.

Cyclosporine Concentration Prediction using Clustering and Support Vector Regression

admin — 2007-01-31T10:09:36Z

www.kernel-machines.org opens

admin — 2007-02-12T09:41:52Z

We are pleased to announce a new website on Kernel Machines and related methods. It is a superset of the Support Vector website at GMD FIRST. Most links to svm.first.gmd.de will still be operational and should result in the near future (as soon as the changes are made to the site in Berlin) in a redirect of your browser to the new site. However, we would like to ask you to update any existing links.

Nips 2004 Workshop on Learning With Structured Outputs

mlang — 2007-07-26T19:19:26Z

Submissions are invited for the workshop on learning with structured outputs, to be held during the NIPS workshops on Dec. 11-13, 2004.

FAQ

admin — 2010-07-12T11:42:19Z

Frequently Asked Questions

Software

admin — 2009-12-14T13:52:18Z

Kernel-Machines.Org software links