Header logo is ei


2008


no image
The role of stimulus correlations for population decoding in the retina

Schwartz, G., Macke, J., Berry, M.

Computational and Systems Neuroscience 2008 (COSYNE 2008), 5, pages: 172, March 2008 (poster)

PDF Web [BibTex]

2008

PDF Web [BibTex]


no image
Causal Reasoning by Evaluating the Complexity of Conditional Densities with Kernel Methods

Sun, X., Janzing, D., Schölkopf, B.

Neurocomputing, 71(7-9):1248-1256, March 2008 (article)

Abstract
We propose a method to quantify the complexity of conditional probability measures by a Hilbert space seminorm of the logarithm of its density. The concept of reproducing kernel Hilbert spaces (RKHSs) is a flexible tool to define such a seminorm by choosing an appropriate kernel. We present several examples with artificial data sets where our kernel-based complexity measure is consistent with our intuitive understanding of complexity of densities. The intention behind the complexity measure is to provide a new approach to inferring causal directions. The idea is that the factorization of the joint probability measure P(effect, cause) into P(effect|cause)P(cause) leads typically to "simpler" and "smoother" terms than the factorization into P(cause|effect)P(effect). Since the conventional constraint-based approach of causal discovery is not able to determine the causal direction between only two variables, our inference principle can in particular be useful when combined with other existing methods. We provide several simple examples with real-world data where the true causal directions indeed lead to simpler (conditional) densities.

Web DOI [BibTex]


no image
Natural Actor-Critic

Peters, J., Schaal, S.

Neurocomputing, 71(7-9):1180-1190, March 2008 (article)

Abstract
In this paper, we suggest a novel reinforcement learning architecture, the Natural Actor-Critic. The actor updates are achieved using stochastic policy gradients em- ploying Amari’s natural gradient approach, while the critic obtains both the natural policy gradient and additional parameters of a value function simultaneously by lin- ear regression. We show that actor improvements with natural policy gradients are particularly appealing as these are independent of coordinate frame of the chosen policy representation, and can be estimated more efficiently than regular policy gra- dients. The critic makes use of a special basis function parameterization motivated by the policy-gradient compatible function approximation. We show that several well-known reinforcement learning methods such as the original Actor-Critic and Bradtke’s Linear Quadratic Q-Learning are in fact Natural Actor-Critic algorithms. Empirical evaluations illustrate the effectiveness of our techniques in comparison to previous methods, and also demonstrate their applicability for learning control on an anthropomorphic robot arm.

PDF PDF DOI [BibTex]

PDF PDF DOI [BibTex]


no image
Poisson Geometry of Parabolic Bundles on Elliptic Curves

Balduzzi, D.

International Journal of Mathematics , 19(3):339-367, March 2008 (article)

Abstract
The moduli space of G-bundles on an elliptic curve with additional flag structure admits a Poisson structure. The bivector can be defined using double loop group, loop group and sheaf cohomology constructions. We investigate the links between these methods and for the case SL2 perform explicit computations, describing the bracket and its leaves in detail.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Inferring Spike Trains From Local Field Potentials

Rasch, M., Gretton, A., Murayama, Y., Maass, W., Logothetis, N.

Journal of Neurophysiology, 99(3):1461-1476, March 2008 (article)

Abstract
We investigated whether it is possible to infer spike trains solely on the basis of the underlying local field potentials (LFPs). Using support vector machines and linear regression models, we found that in the primary visual cortex (V1) of monkeys, spikes can indeed be inferred from LFPs, at least with moderate success. Although there is a considerable degree of variation across electrodes, the low-frequency structure in spike trains (in the 100-ms range) can be inferred with reasonable accuracy, whereas exact spike positions are not reliably predicted. Two kinds of features of the LFP are exploited for prediction: the frequency power of bands in the high gamma-range (40–90 Hz) and information contained in lowfrequency oscillations ( 10 Hz), where both phase and power modulations are informative. Information analysis revealed that both features code (mainly) independent aspects of the spike-to-LFP relationship, with the low-frequency LFP phase coding for temporally clustered spiking activity. Although both features and prediction quality are similar during seminatural movie stimuli and spontaneous activity, prediction performance during spontaneous activity degrades much more slowly with increasing electrode distance. The general trend of data obtained with anesthetized animals is qualitatively mirrored in that of a more limited data set recorded in V1 of non-anesthetized monkeys. In contrast to the cortical field potentials, thalamic LFPs (e.g., LFPs derived from recordings in the dorsal lateral geniculate nucleus) hold no useful information for predicting spiking activity.

Web DOI [BibTex]

Web DOI [BibTex]


no image
ISD: A Software Package for Bayesian NMR Structure Calculation

Rieping, W., Nilges, M., Habeck, M.

Bioinformatics, 24(8):1104-1105, February 2008 (article)

Abstract
SUMMARY: The conventional approach to calculating biomolecular structures from nuclear magnetic resonance (NMR) data is often viewed as subjective due to its dependence on rules of thumb for deriving geometric constraints and suitable values for theory parameters from noisy experimental data. As a result, it can be difficult to judge the precision of an NMR structure in an objective manner. The Inferential Structure Determination (ISD) framework, which has been introduced recently, addresses this problem by using Bayesian inference to derive a probability distribution that represents both the unknown structure and its uncertainty. It also determines additional unknowns, such as theory parameters, that normally need be chosen empirically. Here we give an overview of the ISD software package, which implements this methodology. AVAILABILITY: The program is available at http://www.bioc.cam.ac.uk/isd

Web DOI [BibTex]

Web DOI [BibTex]


no image
Probabilistic Structure Calculation

Nilges, M., Habeck, M., Rieping, W.

Comptes Rendus Chimie, 11(4-5):356-369, February 2008 (article)

Abstract
Molecular structures are usually calculated from experimental data with some method of energy minimisation or non-linear optimisation. Key aims of a structure calculation are to estimate the coordinate uncertainty, and to provide a meaningful measure of the quality of the fit to the data. We discuss approaches to optimally combine prior information and experimental data and the connection to probability theory. We analyse the appropriate statistics for NOEs and NOE-derived distances, and the related question of restraint potentials. Finally, we will discuss approaches to determine the appropriate weight on the experimental evidence and to obtain in this way an estimate of the data quality from the structure calculation. Whereas objective estimates of coordinates and their uncertainties can only be obtained by a full Bayesian treatment of the problem, standard structure calculation methods continue to play an important role. To obtain the full benefit of these methods, they should be founded on a rigorous Baye sian analysis.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Optimization Techniques for Semi-Supervised Support Vector Machines

Chapelle, O., Sindhwani, V., Keerthi, S.

Journal of Machine Learning Research, 9, pages: 203-233, February 2008 (article)

Abstract
Due to its wide applicability, the problem of semi-supervised classification is attracting increasing attention in machine learning. Semi-Supervised Support Vector Machines (S3VMs) are based on applying the margin maximization principle to both labeled and unlabeled examples. Unlike SVMs, their formulation leads to a non-convex optimization problem. A suite of algorithms have recently been proposed for solving S3VMs. This paper reviews key ideas in this literature. The performance and behavior of various S3VMs algorithms is studied together, under a common experimental setting.

PDF [BibTex]

PDF [BibTex]


no image
Fast Projection-based Methods for the Least Squares Nonnegative Matrix Approximation Problem

Kim, D., Sra, S., Dhillon, I.

Statistical Analysis and Data Mining, 1(1):38-51, February 2008 (article)

Abstract
Nonnegative matrix approximation (NNMA) is a popular matrix decomposition technique that has proven to be useful across a diverse variety of fields with applications ranging from document analysis and image processing to bioinformatics and signal processing. Over the years, several algorithms for NNMA have been proposed, e.g. Lee and Seung‘s multiplicative updates, alternating least squares (ALS), and gradient descent-based procedures. However, most of these procedures suffer from either slow convergence, numerical instability, or at worst, serious theoretical drawbacks. In this paper, we develop a new and improved algorithmic framework for the least-squares NNMA problem, which is not only theoretically well-founded, but also overcomes many deficiencies of other methods. Our framework readily admits powerful optimization techniques and as concrete realizations we present implementations based on the Newton, BFGS and conjugate gradient methods. Our algorithms provide numerical resu lts supe rior to both Lee and Seung‘s method as well as to the alternating least squares heuristic, which was reported to work well in some situations but has no theoretical guarantees[1]. Our approach extends naturally to include regularization and box-constraints without sacrificing convergence guarantees. We present experimental results on both synthetic and real-world datasets that demonstrate the superiority of our methods, both in terms of better approximations as well as computational efficiency.

Web DOI [BibTex]

Web DOI [BibTex]


no image
A Unifying Probabilistic Framework for Analyzing Residual Dipolar Couplings

Habeck, M., Nilges, M., Rieping, W.

Journal of Biomolecular NMR, 40(2):135-144, February 2008 (article)

Abstract
Residual dipolar couplings provide complementary information to the nuclear Overhauser effect measurements that are traditionally used in biomolecular structure determination by NMR. In a de novo structure determination, however, lack of knowledge about the degree and orientation of molecular alignment complicates the analysis of dipolar coupling data. We present a probabilistic framework for analyzing residual dipolar couplings and demonstrate that it is possible to estimate the atomic coordinates, the complete molecular alignment tensor, and the error of the couplings simultaneously. As a by-product, we also obtain estimates of the uncertainty in the coordinates and the alignment tensor. We show that our approach encompasses existing methods for determining the alignment tensor as special cases, including least squares estimation, histogram fitting, and elimination of an explicit alignment tensor in the restraint energy.

PDF DOI [BibTex]

PDF DOI [BibTex]


no image
Development and Application of a Python Scripting Framework for BCI2000

Schreiner, T.

Biologische Kybernetik, Eberhard-Karls-Universität Tübingen, Tübingen, Germany, January 2008 (diplomathesis)

[BibTex]

[BibTex]


no image
Efficient and Invariant Regularisation with Application to Computer Graphics

Walder, CJ.

Biologische Kybernetik, University of Queensland, Brisbane, Australia, January 2008 (phdthesis)

Abstract
This thesis develops the theory and practise of reproducing kernel methods. Many functional inverse problems which arise in, for example, machine learning and computer graphics, have been treated with practical success using methods based on a reproducing kernel Hilbert space perspective. This perspective is often theoretically convenient, in that many functional analysis problems reduce to linear algebra problems in these spaces. Somewhat more complex is the case of conditionally positive definite kernels, and we provide an introduction to both cases, deriving in a particularly elementary manner some key results for the conditionally positive definite case. A common complaint of the practitioner is the long running time of these kernel based algorithms. We provide novel ways of alleviating these problems by essentially using a non-standard function basis which yields computational advantages. That said, by doing so we must also forego the aforementioned theoretical conveniences, and hence need some additional analysis which we provide in order to make the approach practicable. We demonstrate that the method leads to state of the art performance on the problem of surface reconstruction from points. We also provide some analysis of kernels invariant to transformations such as translation and dilation, and show that this indicates the value of learning algorithms which use conditionally positive definite kernels. Correspondingly, we provide a few approaches for making such algorithms practicable. We do this either by modifying the kernel, or directly solving problems with conditionally positive definite kernels, which had previously only been solved with positive definite kernels. We demonstrate the advantage of this approach, in particular by attaining state of the art classification performance with only one free parameter.

PDF [BibTex]

PDF [BibTex]


no image
Contour-propagation Algorithms for Semi-automated Reconstruction of Neural Processes

Macke, J., Maack, N., Gupta, R., Denk, W., Schölkopf, B., Borst, A.

Journal of Neuroscience Methods, 167(2):349-357, January 2008 (article)

Abstract
A new technique, ”Serial Block Face Scanning Electron Microscopy” (SBFSEM), allows for automatic sectioning and imaging of biological tissue with a scanning electron microscope. Image stacks generated with this technology have a resolution sufficient to distinguish different cellular compartments, including synaptic structures, which should make it possible to obtain detailed anatomical knowledge of complete neuronal circuits. Such an image stack contains several thousands of images and is recorded with a minimal voxel size of 10-20nm in the x and y- and 30nm in z-direction. Consequently, a tissue block of 1mm3 (the approximate volume of the Calliphora vicina brain) will produce several hundred terabytes of data. Therefore, highly automated 3D reconstruction algorithms are needed. As a first step in this direction we have developed semiautomated segmentation algorithms for a precise contour tracing of cell membranes. These algorithms were embedded into an easy-to-operate user interface, which allows direct 3D observation of the extracted objects during the segmentation of image stacks. Compared to purely manual tracing, processing time is greatly accelerated.

PDF Web DOI [BibTex]

PDF Web DOI [BibTex]


no image
A Quantum-Statistical-Mechanical Extension of Gaussian Mixture Model

Tanaka, K., Tsuda, K.

Journal of Physics: Conference Series, 95(012023):1-9, January 2008 (article)

Abstract
We propose an extension of Gaussian mixture models in the statistical-mechanical point of view. The conventional Gaussian mixture models are formulated to divide all points in given data to some kinds of classes. We introduce some quantum states constructed by superposing conventional classes in linear combinations. Our extension can provide a new algorithm in classifications of data by means of linear response formulas in the statistical mechanics.

PDF PDF DOI [BibTex]

PDF PDF DOI [BibTex]


no image
Energy Functionals for Manifold-valued Mappings and Their Properties

Hein, M., Steinke, F., Schölkopf, B.

(167), Max Planck Institute for Biological Cybernetics, Tübingen, January 2008 (techreport)

Abstract
This technical report is merely an extended version of the appendix of Steinke et.al. "Manifold-valued Thin-Plate Splines with Applications in Computer Graphics" (2008) with complete proofs, which had to be omitted due to space restrictions. This technical report requires a basic knowledge of differential geometry. However, apart from that requirement the technical report is self-contained.

PDF [BibTex]

PDF [BibTex]


no image
Methods for feature selection in a learning machine

Weston, J., Elisseeff, A., Schölkopf, B., Pérez-Cruz, F.

United States Patent, No 7318051, January 2008 (patent)

[BibTex]

[BibTex]


no image
A Robot System for Biomimetic Navigation: From Snapshots to Metric Embeddings of View Graphs

Franz, MO., Stürzl, W., Reichardt, W., Mallot, HA.

In Robotics and Cognitive Approaches to Spatial Mapping, pages: 297-314, Springer Tracts in Advanced Robotics ; 38, (Editors: Jefferies, M.E. , W.-K. Yeap), Springer, Berlin, Germany, 2008 (inbook)

Abstract
Complex navigation behaviour (way-finding) involves recognizing several places and encoding a spatial relationship between them. Way-finding skills can be classified into a hierarchy according to the complexity of the tasks that can be performed [8]. The most basic form of way-finding is route navigation, followed by topological navigation where several routes are integrated into a graph-like representation. The highest level, survey navigation, is reached when this graph can be embedded into a common reference frame. In this chapter, we present the building blocks for a biomimetic robot navigation system that encompasses all levels of this hierarchy. As a local navigation method, we use scene-based homing. In this scheme, a goal location is characterized either by a panoramic snapshot of the light intensities as seen from the place, or by a record of the distances to the surrounding objects. The goal is found by moving in the direction that minimizes the discrepancy between the recorded intensities or distances and the current sensory input. For learning routes, the robot selects distinct views during exploration that are close enough to be reached by snapshot-based homing. When it encounters already visited places during route learning, it connects the routes and thus forms a topological representation of its environment termed a view graph. The final stage, survey navigation, is achieved by a graph embedding procedure which complements the topologic information of the view graph with odometric position estimates. Calculation of the graph embedding is done with a modified multidimensional scaling algorithm which makes use of distances and angles between nodes.

PDF PDF DOI [BibTex]

PDF PDF DOI [BibTex]


no image
Haptic Device For Cell Manipulation

Lee, DY., Son, HI., Woo, HJ.

Max-Planck-Gesellschaft, Biologische Kybernetik, 2008 (patent)

[BibTex]

[BibTex]


no image
Minimal Nonlinear Distortion Principle for Nonlinear Independent Component Analysis

Zhang, K., Chan, L.

Journal of Machine Learning Research, 9, pages: 2455-2487, 2008 (article)

Abstract
It is well known that solutions to the nonlinear independent component analysis (ICA) problem are highly non-unique. In this paper we propose the "minimal nonlinear distortion" (MND) principle for tackling the ill-posedness of nonlinear ICA problems. MND prefers the nonlinear ICA solution with the estimated mixing procedure as close as possible to linear, among all possible solutions. It also helps to avoid local optima in the solutions. To achieve MND, we exploit a regularization term to minimize the mean square error between the nonlinear mixing mapping and the best-fitting linear one. The effect of MND on the inherent trivial and non-trivial indeterminacies in nonlinear ICA solutions is investigated. Moreover, we show that local MND is closely related to the smoothness regularizer penalizing large curvature, which provides another useful regularization condition for nonlinear ICA. Experiments on synthetic data show the usefulness of the MND principle for separating various nonlinear mixtures. Finally, as an application, we use nonlinear ICA with MND to separate daily returns of a set of stocks in Hong Kong, and the linear causal relations among them are successfully discovered. The resulting causal relations give some interesting insights into the stock market. Such a result can not be achieved by linear ICA. Simulation studies also verify that when doing causality discovery, sometimes one should not ignore the nonlinear distortion in the data generation procedure, even if it is weak.

Web [BibTex]

Web [BibTex]


no image
Transport processes in networks with scattering ramification nodes

Radl, A.

Journal of Applied Functional Analysis, 3, pages: 461-483, 2008 (article)

Web [BibTex]

Web [BibTex]


no image
Learning to control in operational space

Peters, J., Schaal, S.

International Journal of Robotics Research, 27, pages: 197-212, 2008, clmc (article)

Abstract
One of the most general frameworks for phrasing control problems for complex, redundant robots is operational space control. However, while this framework is of essential importance for robotics and well-understood from an analytical point of view, it can be prohibitively hard to achieve accurate control in face of modeling errors, which are inevitable in com- plex robots, e.g., humanoid robots. In this paper, we suggest a learning approach for opertional space control as a direct inverse model learning problem. A first important insight for this paper is that a physically cor- rect solution to the inverse problem with redundant degrees-of-freedom does exist when learning of the inverse map is performed in a suitable piecewise linear way. The second crucial component for our work is based on the insight that many operational space controllers can be understood in terms of a constrained optimal control problem. The cost function as- sociated with this optimal control problem allows us to formulate a learn- ing algorithm that automatically synthesizes a globally consistent desired resolution of redundancy while learning the operational space controller. From the machine learning point of view, this learning problem corre- sponds to a reinforcement learning problem that maximizes an immediate reward. We employ an expectation-maximization policy search algorithm in order to solve this problem. Evaluations on a three degrees of freedom robot arm are used to illustrate the suggested approach. The applica- tion to a physically realistic simulator of the anthropomorphic SARCOS Master arm demonstrates feasibility for complex high degree-of-freedom robots. We also show that the proposed method works in the setting of learning resolved motion rate control on real, physical Mitsubishi PA-10 medical robotics arm.

link (url) DOI [BibTex]

link (url) DOI [BibTex]

2000


no image
Knowledge Discovery in Databases: An Information Retrieval Perspective

Ong, CS.

Malaysian Journal of Computer Science, 13(2):54-63, December 2000 (article)

Abstract
The current trend of increasing capabilities in data generation and collection has resulted in an urgent need for data mining applications, also called knowledge discovery in databases. This paper identifies and examines the issues involved in extracting useful grains of knowledge from large amounts of data. It describes a framework to categorise data mining systems. The author also gives an overview of the issues pertaining to data pre processing, as well as various information gathering methodologies and techniques. The paper covers some popular tools such as classification, clustering, and generalisation. A summary of statistical and machine learning techniques used currently is also provided.

PDF [BibTex]

2000

PDF [BibTex]


no image
A real-time model of the human knee for application in virtual orthopaedic trainer

Peters, J., Riener, R.

In Proceedings of the 10th International Conference on BioMedical Engineering (ICBME 2000), 10, pages: 1-2, 10th International Conference on BioMedical Engineering (ICBME) , December 2000 (inproceedings)

Abstract
In this paper a real-time capable computational model of the human knee is presented. The model describes the passive elastic joint characteristics in six degrees-of-freedom (DOF). A black-box approach was chosen, where experimental data were approximated by piecewise polynomial functions. The knee model has been applied in a the Virtual Orthopaedic Trainer, which can support training of physical knee evaluation required for diagnosis and surgical planning.

PDF Web [BibTex]

PDF Web [BibTex]


no image
A Simple Iterative Approach to Parameter Optimization

Zien, A., Zimmer, R., Lengauer, T.

Journal of Computational Biology, 7(3,4):483-501, November 2000 (article)

Abstract
Various bioinformatics problems require optimizing several different properties simultaneously. For example, in the protein threading problem, a scoring function combines the values for different parameters of possible sequence-to-structure alignments into a single score to allow for unambiguous optimization. In this context, an essential question is how each property should be weighted. As the native structures are known for some sequences, a partial ordering on optimal alignments to other structures, e.g., derived from structural comparisons, may be used to adjust the weights. To resolve the arising interdependence of weights and computed solutions, we propose a heuristic approach: iterating the computation of solutions (here, threading alignments) given the weights and the estimation of optimal weights of the scoring function given these solutions via systematic calibration methods. For our application (i.e., threading), this iterative approach results in structurally meaningful weights that significantly improve performance on both the training and the test data sets. In addition, the optimized parameters show significant improvements on the recognition rate for a grossly enlarged comprehensive benchmark, a modified recognition protocol as well as modified alignment types (local instead of global and profiles instead of single sequences). These results show the general validity of the optimized weights for the given threading program and the associated scoring contributions.

Web [BibTex]

Web [BibTex]


no image
Identification of Drug Target Proteins

Zien, A., Küffner, R., Mevissen, T., Zimmer, R., Lengauer, T.

ERCIM News, 43, pages: 16-17, October 2000 (article)

Web [BibTex]

Web [BibTex]


no image
On Designing an Automated Malaysian Stemmer for the Malay Language

Tai, SY., Ong, CS., Abullah, NA.

In Fifth International Workshop on Information Retrieval with Asian Languages, pages: 207-208, ACM Press, New York, NY, USA, Fifth International Workshop on Information Retrieval with Asian Languages, October 2000 (inproceedings)

Abstract
Online and interactive information retrieval systems are likely to play an increasing role in the Malay Language community. To facilitate and automate the process of matching morphological term variants, a stemmer focusing on common affix removal algorithms is proposed as part of the design of an information retrieval system for the Malay Language. Stemming is a morphological process of normalizing word tokens down to their essential roots. The proposed stemmer strips prefixes and suffixes off the word. The experiment conducted with web sites selected from the World Wide Web has exhibited substantial improvements in the number of words indexed.

PostScript Web DOI [BibTex]

PostScript Web DOI [BibTex]


no image
Robust ensemble learning

Rätsch, G., Schölkopf, B., Smola, A., Mika, S., Onoda, T., Müller, K.

In Advances in Large Margin Classifiers, pages: 207-220, Neural Information Processing Series, (Editors: AJ Smola and PJ Bartlett and B Schölkopf and D. Schuurmans), MIT Press, Cambridge, MA, USA, October 2000 (inbook)

[BibTex]

[BibTex]


no image
Entropy numbers for convex combinations and MLPs

Smola, A., Elisseeff, A., Schölkopf, B., Williamson, R.

In Advances in Large Margin Classifiers, pages: 369-387, Neural Information Processing Series, (Editors: AJ Smola and PL Bartlett and B Schölkopf and D Schuurmans), MIT Press, Cambridge, MA,, October 2000 (inbook)

[BibTex]

[BibTex]


no image
Ensemble of Specialized Networks based on Input Space Partition

Shin, H., Lee, H., Cho, S.

In Proc. of the Korean Operations Research and Management Science Conference, pages: 33-36, Korean Operations Research and Management Science Conference, October 2000 (inproceedings)

[BibTex]

[BibTex]


no image
DES Approach Failure Recovery of Pump-valve System

Son, HI., Kim, KW., Lee, S.

In Korean Society of Precision Engineering (KSPE) Conference, pages: 647-650, Annual Meeting of the Korean Society of Precision Engineering (KSPE), October 2000 (inproceedings)

PDF [BibTex]

PDF [BibTex]


no image
Natural Regularization from Generative Models

Oliver, N., Schölkopf, B., Smola, A.

In Advances in Large Margin Classifiers, pages: 51-60, Neural Information Processing Series, (Editors: AJ Smola and PJ Bartlett and B Schölkopf and D Schuurmans), MIT Press, Cambridge, MA, USA, October 2000 (inbook)

[BibTex]

[BibTex]


no image
Advances in Large Margin Classifiers

Smola, A., Bartlett, P., Schölkopf, B., Schuurmans, D.

pages: 422, Neural Information Processing, MIT Press, Cambridge, MA, USA, October 2000 (book)

Abstract
The concept of large margins is a unifying principle for the analysis of many different approaches to the classification of data from examples, including boosting, mathematical programming, neural networks, and support vector machines. The fact that it is the margin, or confidence level, of a classification--that is, a scale parameter--rather than a raw training error that matters has become a key tool for dealing with classifiers. This book shows how this idea applies to both the theoretical analysis and the design of algorithms. The book provides an overview of recent developments in large margin classifiers, examines connections with other methods (e.g., Bayesian inference), and identifies strengths and weaknesses of the method, as well as directions for future research. Among the contributors are Manfred Opper, Vladimir Vapnik, and Grace Wahba.

Web [BibTex]

Web [BibTex]


no image
Ensemble Learning Algorithm of Specialized Networks

Shin, H., Lee, H., Cho, S.

In Proc. of the Korea Information Science Conference, pages: 308-310, Korea Information Science Conference, October 2000 (inproceedings)

[BibTex]

[BibTex]


no image
DES Approach Failure Diagnosis of Pump-valve System

Son, HI., Kim, KW., Lee, S.

In Korean Society of Precision Engineering (KSPE) Conference, pages: 643-646, Annual Meeting of the Korean Society of Precision Engineering (KSPE), October 2000 (inproceedings)

Abstract
As many industrial systems become more complex, it becomes extremely difficult to diagnose the cause of failures. This paper presents a failure diagnosis approach based on discrete event system theory. In particular, the approach is a hybrid of event-based and state-based ones leading to a simpler failure diagnoser with supervisory control capability. The design procedure is presented along with a pump-valve system as an example.

PDF [BibTex]

PDF [BibTex]


no image
Engineering Support Vector Machine Kernels That Recognize Translation Initiation Sites

Zien, A., Rätsch, G., Mika, S., Schölkopf, B., Lengauer, T., Müller, K.

Bioinformatics, 16(9):799-807, September 2000 (article)

Abstract
Motivation: In order to extract protein sequences from nucleotide sequences, it is an important step to recognize points at which regions start that code for proteins. These points are called translation initiation sites (TIS). Results: The task of finding TIS can be modeled as a classification problem. We demonstrate the applicability of support vector machines for this task, and show how to incorporate prior biological knowledge by engineering an appropriate kernel function. With the described techniques the recognition performance can be improved by 26% over leading existing approaches. We provide evidence that existing related methods (e.g. ESTScan) could profit from advanced TIS recognition.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Three-dimensional reconstruction of planar scenes

Urbanek, M.

Biologische Kybernetik, INP Grenoble, Warsaw University of Technology, September 2000 (diplomathesis)

Abstract
For a planar scene, we propose an algorithm to estimate its 3D structure. Homographies between corresponding planes are employed in order to recover camera motion parameters - between camera positions from which images of the scene were taken. Cases of one- and multiple- corresponding planes present on the scene are distinguished. Solutions are proposed for both cases.

ZIP [BibTex]

ZIP [BibTex]


no image
Analysis of Gene Expression Data with Pathway Scores

Zien, A., Küffner, R., Zimmer, R., Lengauer, T.

In ISMB 2000, pages: 407-417, AAAI Press, Menlo Park, CA, USA, 8th International Conference on Intelligent Systems for Molecular Biology, August 2000 (inproceedings)

Abstract
We present a new approach for the evaluation of gene expression data. The basic idea is to generate biologically possible pathways and to score them with respect to gene expression measurements. We suggest sample scoring functions for different problem specifications. The significance of the scores for the investigated pathways is assessed by comparison to a number of scores for random pathways. We show that simple scoring functions can assign statistically significant scores to biologically relevant pathways. This suggests that the combination of appropriate scoring functions with the systematic generation of pathways can be used in order to select the most interesting pathways based on gene expression measurements.

PDF [BibTex]

PDF [BibTex]


no image
A Meanfield Approach to the Thermodynamics of a Protein-Solvent System with Application to the Oligomerization of the Tumour Suppressor p53.

Noolandi, J., Davison, TS., Vokel, A., Nie, F., Kay, C., Arrowsmith, C.

Proceedings of the National Academy of Sciences of the United States of America, 97(18):9955-9960, August 2000 (article)

Web [BibTex]

Web [BibTex]


no image
Observational Learning with Modular Networks

Shin, H., Lee, H., Cho, S.

In Lecture Notes in Computer Science (LNCS 1983), LNCS 1983, pages: 126-132, Springer-Verlag, Heidelberg, International Conference on Intelligent Data Engineering and Automated Learning (IDEAL), July 2000 (inproceedings)

Abstract
Observational learning algorithm is an ensemble algorithm where each network is initially trained with a bootstrapped data set and virtual data are generated from the ensemble for training. Here we propose a modular OLA approach where the original training set is partitioned into clusters and then each network is instead trained with one of the clusters. Networks are combined with different weighting factors now that are inversely proportional to the distance from the input vector to the cluster centers. Comparison with bagging and boosting shows that the proposed approach reduces generalization error with a smaller number of networks employed.

PDF [BibTex]

PDF [BibTex]


no image
The Infinite Gaussian Mixture Model

Rasmussen, CE.

In Advances in Neural Information Processing Systems 12, pages: 554-560, (Editors: Solla, S.A. , T.K. Leen, K-R Müller), MIT Press, Cambridge, MA, USA, Thirteenth Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

Abstract
In a Bayesian mixture model it is not necessary a priori to limit the number of components to be finite. In this paper an infinite Gaussian mixture model is presented which neatly sidesteps the difficult problem of finding the ``right'' number of mixture components. Inference in the model is done using an efficient parameter-free Markov Chain that relies entirely on Gibbs sampling.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Generalization Abilities of Ensemble Learning Algorithms

Shin, H., Jang, M., Cho, S.

In Proc. of the Korean Brain Society Conference, pages: 129-133, Korean Brain Society Conference, June 2000 (inproceedings)

[BibTex]

[BibTex]


no image
Support vector method for novelty detection

Schölkopf, B., Williamson, R., Smola, A., Shawe-Taylor, J., Platt, J.

In Advances in Neural Information Processing Systems 12, pages: 582-588, (Editors: SA Solla and TK Leen and K-R Müller), MIT Press, Cambridge, MA, USA, 13th Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

Abstract
Suppose you are given some dataset drawn from an underlying probability distribution ¤ and you want to estimate a “simple” subset ¥ of input space such that the probability that a test point drawn from ¤ lies outside of ¥ equals some a priori specified ¦ between § and ¨. We propose a method to approach this problem by trying to estimate a function © which is positive on ¥ and negative on the complement. The functional form of © is given by a kernel expansion in terms of a potentially small subset of the training data; it is regularized by controlling the length of the weight vector in an associated feature space. We provide a theoretical analysis of the statistical performance of our algorithm. The algorithm is a natural extension of the support vector algorithm to the case of unlabelled data.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Solving Satisfiability Problems with Genetic Algorithms

Harmeling, S.

In Genetic Algorithms and Genetic Programming at Stanford 2000, pages: 206-213, (Editors: Koza, J. R.), Stanford Bookstore, Stanford, CA, USA, June 2000 (inbook)

Abstract
We show how to solve hard 3-SAT problems using genetic algorithms. Furthermore, we explore other genetic operators that may be useful to tackle 3-SAT problems, and discuss their pros and cons.

PDF [BibTex]

PDF [BibTex]


no image
v-Arc: Ensemble Learning in the Presence of Outliers

Rätsch, G., Schölkopf, B., Smola, A., Müller, K., Onoda, T., Mika, S.

In Advances in Neural Information Processing Systems 12, pages: 561-567, (Editors: SA Solla and TK Leen and K-R Müller), MIT Press, Cambridge, MA, USA, 13th Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

Abstract
AdaBoost and other ensemble methods have successfully been applied to a number of classification tasks, seemingly defying problems of overfitting. AdaBoost performs gradient descent in an error function with respect to the margin, asymptotically concentrating on the patterns which are hardest to learn. For very noisy problems, however, this can be disadvantageous. Indeed, theoretical analysis has shown that the margin distribution, as opposed to just the minimal margin, plays a crucial role in understanding this phenomenon. Loosely speaking, some outliers should be tolerated if this has the benefit of substantially increasing the margin on the remaining points. We propose a new boosting algorithm which allows for the possibility of a pre-specified fraction of points to lie in the margin area or even on the wrong side of the decision boundary.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Invariant feature extraction and classification in kernel spaces

Mika, S., Rätsch, G., Weston, J., Schölkopf, B., Smola, A., Müller, K.

In Advances in neural information processing systems 12, pages: 526-532, (Editors: SA Solla and TK Leen and K-R Müller), MIT Press, Cambridge, MA, USA, 13th Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

PDF Web [BibTex]

PDF Web [BibTex]


no image
Transductive Inference for Estimating Values of Functions

Chapelle, O., Vapnik, V., Weston, J.

In Advances in Neural Information Processing Systems 12, pages: 421-427, (Editors: Solla, S.A. , T.K. Leen, K-R Müller), MIT Press, Cambridge, MA, USA, Thirteenth Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

Abstract
We introduce an algorithm for estimating the values of a function at a set of test points $x_1^*,dots,x^*_m$ given a set of training points $(x_1,y_1),dots,(x_ell,y_ell)$ without estimating (as an intermediate step) the regression function. We demonstrate that this direct (transductive) way for estimating values of the regression (or classification in pattern recognition) is more accurate than the traditional one based on two steps, first estimating the function and then calculating the values of this function at the points of interest.

PDF Web [BibTex]

PDF Web [BibTex]


no image
The entropy regularization information criterion

Smola, A., Shawe-Taylor, J., Schölkopf, B., Williamson, R.

In Advances in Neural Information Processing Systems 12, pages: 342-348, (Editors: SA Solla and TK Leen and K-R Müller), MIT Press, Cambridge, MA, USA, 13th Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

Abstract
Effective methods of capacity control via uniform convergence bounds for function expansions have been largely limited to Support Vector machines, where good bounds are obtainable by the entropy number approach. We extend these methods to systems with expansions in terms of arbitrary (parametrized) basis functions and a wide range of regularization methods covering the whole range of general linear additive models. This is achieved by a data dependent analysis of the eigenvalues of the corresponding design matrix.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Model Selection for Support Vector Machines

Chapelle, O., Vapnik, V.

In Advances in Neural Information Processing Systems 12, pages: 230-236, (Editors: Solla, S.A. , T.K. Leen, K-R Müller), MIT Press, Cambridge, MA, USA, Thirteenth Annual Neural Information Processing Systems Conference (NIPS), June 2000 (inproceedings)

Abstract
New functionals for parameter (model) selection of Support Vector Machines are introduced based on the concepts of the span of support vectors and rescaling of the feature space. It is shown that using these functionals, one can both predict the best choice of parameters of the model and the relative quality of performance for any value of parameter.

PDF Web [BibTex]

PDF Web [BibTex]


no image
New Support Vector Algorithms

Schölkopf, B., Smola, A., Williamson, R., Bartlett, P.

Neural Computation, 12(5):1207-1245, May 2000 (article)

Abstract
We propose a new class of support vector algorithms for regression and classification. In these algorithms, a parameter {nu} lets one effectively control the number of support vectors. While this can be useful in its own right, the parameterization has the additional benefit of enabling us to eliminate one of the other free parameters of the algorithm: the accuracy parameter {epsilon} in the regression case, and the regularization constant C in the classification case. We describe the algorithms, give some theoretical results concerning the meaning and the choice of {nu}, and report experimental results.

Web DOI [BibTex]

Web DOI [BibTex]


no image
Generalization Abilities of Ensemble Learning Algorithms: OLA, Bagging, Boosting

Shin, H., Jang, M., Cho, S., Lee, B., Lim, Y.

In Proc. of the Korea Information Science Conference, pages: 226-228, Conference on Korean Information Science, April 2000 (inproceedings)

[BibTex]

[BibTex]