Applications of Kernel Machines to Structured Data

Eichhorn, J

Local TagsRelease HistoryDetailsSummary

Applications of Kernel Machines to Structured Data

Eichhorn, J. (2007). Applications of Kernel Machines to Structured Data. PhD Thesis, Technische Universität Berlin, Berlin, Germany.

Item is Released

show all hide all

Basic

show hide

Item Permalink: https://hdl.handle.net/11858/00-001M-0000-0013-CE93-9 Version Permalink: https://hdl.handle.net/21.11116/0000-0006-C6DB-4

Genre: Thesis

Files

show Files

Locators

show

hide

Locator:
https://depositonce.tu-berlin.de/handle/11303/1844 (Publisher version) Open Access status unknown

Description:
-

OA-Status:

Creators

show

hide

Creators:
Eichhorn, J^{1, 2}, Author

Affiliations:
1Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795
2Max Planck Institute for Biological Cybernetics, Max Planck Society, Spemannstrasse 38, 72076 Tübingen, DE, ou_1497794

Content

show

hide

Free keywords: -

Abstract: In this thesis we are concerned with the application of supervised learning methods to two problems of rather different nature -- one originating from computational neuroscience, the other one from computer vision. The kernel algorithms that will be used allow classification of complex objects that need not to be elements of a Euclidean vector space. For example in the applications presented below these objects are time series of neural activity and images described by a collection of local descriptors. The flexibility of kernel algorithms is achieved through the use of a kernel function that specifies similarity of the objects as a numerical value. To make an application successful, one has to find appropriate kernel functions that adequately describe similarity and at the same time must fulfil certain mathematical requirements. The focus of our work is the development and adaptation of kernel functions for decoding of neural activity and for image categorisation. Each topic is treated separately in one of the two parts of this thesis. In part~\ref{part:neuro} the application of kernel algorithms for decoding of neural activity is explored. Sequences of action potentials that were measured as response to a visual stimulus are used to reconstruct characteristic attributes of the stimulus. Most of the current methods in neuroscience consider only the number of action potentials in a certain time interval and neglect the temporal distribution of these events. With the kernel functions for neural activity that are proposed in this thesis an extended analysis of spike trains is possible. The similarity of two sequences is not only determined by the frequency of spikes but also takes potential temporal patterns into account. An evaluation of the kernels is performed on artificially generated data as well as on real recordings from a neurophysiological experiment. Experiments on this second type of data allow some conclusions about the actual importance of temporal patterns for the encoding of stimulus attributes in the organism under consideration. In a second set of experiments the simultaneously recorded activity of multiple neurons is taken as a basis of reconstruction. Here the results show that the tested kernel algorithms can perform reconstruction in most cases with a significantly higher precision than current methods of computational neuroscience. The second part of this thesis presents an application of support vector machines as one prominent example of kernel algorithms to the task of object categorisation. Computer vision research has found that it is advantageous for many problems to represent images as a collection of image parts. These parts describe bounded regions of the image that have been previously selected for being particularly salient. To define a kernel function on this type of image representation, we neglected geometrical relations among the image parts and applied two recently proposed kernel functions for sets. The usefulness of this approach was tested on two standard data-sets for image categorisation and was compared to other methods when taking part in an open challenge on visual object categorisation. Results of the challenge show that the use of support vector machines in object categorisation can provide a substantial advantage in performance.

Details

show

hide

Language(s):

Dates: Published Online: 2007-03-05Date issued: 2007

Publication Status: Issued

Pages: 157

Publishing info: Berlin, Germany : Technische Universität Berlin

Table of Contents: -

Rev. Type: -

Identifiers: BibTex Citekey: 4421

Degree: PhD

Event

show

Legal Case

show

Project information

show

Source

show