Header logo is ei


1996


no image
Incorporating invariances in support vector learning machines

Schölkopf, B., Burges, C., Vapnik, V.

In Artificial Neural Networks: ICANN 96, LNCS vol. 1112, pages: 47-52, (Editors: C von der Malsburg and W von Seelen and JC Vorbrüggen and B Sendhoff), Springer, Berlin, Germany, 6th International Conference on Artificial Neural Networks, July 1996, volume 1112 of Lecture Notes in Computer Science (inproceedings)

Abstract
Developed only recently, support vector learning machines achieve high generalization ability by minimizing a bound on the expected test error; however, so far there existed no way of adding knowledge about invariances of a classification problem at hand. We present a method of incorporating prior knowledge about transformation invariances by applying transformations to support vectors, the training examples most critical for determining the classification boundary.

PDF DOI [BibTex]

1996

PDF DOI [BibTex]


no image
A practical Monte Carlo implementation of Bayesian learning

Rasmussen, CE.

In Advances in Neural Information Processing Systems 8, pages: 598-604, (Editors: Touretzky, D.S. , M.C. Mozer, M.E. Hasselmo), MIT Press, Cambridge, MA, USA, Ninth Annual Conference on Neural Information Processing Systems (NIPS), June 1996 (inproceedings)

Abstract
A practical method for Bayesian training of feed-forward neural networks using sophisticated Monte Carlo methods is presented and evaluated. In reasonably small amounts of computer time this approach outperforms other state-of-the-art methods on 5 datalimited tasks from real world domains.

PDF Web [BibTex]

PDF Web [BibTex]


no image
Gaussian Processes for Regression

Williams, CKI., Rasmussen, CE.

In Advances in neural information processing systems 8, pages: 514-520, (Editors: Touretzky, D.S. , M.C. Mozer, M.E. Hasselmo), MIT Press, Cambridge, MA, USA, Ninth Annual Conference on Neural Information Processing Systems (NIPS), June 1996 (inproceedings)

Abstract
The Bayesian analysis of neural networks is difficult because a simple prior over weights implies a complex prior over functions. We investigate the use of a Gaussian process prior over functions, which permits the predictive Bayesian analysis for fixed values of hyperparameters to be carried out exactly using matrix operations. Two methods, using optimization and averaging (via Hybrid Monte Carlo) over hyperparameters have been tested on a number of challenging problems and have produced excellent results.

PDF Web [BibTex]

PDF Web [BibTex]

1995


no image
View-based cognitive map learning by an autonomous robot

Mallot, H., Bülthoff, H., Georg, P., Schölkopf, B., Yasuhara, K.

In Proceedings International Conference on Artificial Neural Networks, vol. 2, pages: 381-386, (Editors: Fogelman-Soulié, F.), EC2, Paris, France, Conférence Internationale sur les Réseaux de Neurones Artificiels (ICANN '95), October 1995 (inproceedings)

Abstract
This paper presents a view-based approach to map learning and navigation in mazes. By means of graph theory we have shown that the view-graph is a sufficient representation for map behaviour such as path planning. A neural network for unsupervised learning of the view-graph from sequences of views is constructed. We use a modified Kohonen (1988) learning rule that transforms temporal sequence (rather than featural similarity) into connectedness. In the main part of the paper, we present a robot implementation of the scheme. The results show that the proposed network is able to support map behaviour in simple environments.

PDF [BibTex]

1995

PDF [BibTex]


no image
Extracting support data for a given task

Schölkopf, B., Burges, C., Vapnik, V.

In First International Conference on Knowledge Discovery & Data Mining (KDD-95), pages: 252-257, (Editors: UM Fayyad and R Uthurusamy), AAAI Press, Menlo Park, CA, USA, August 1995 (inproceedings)

Abstract
We report a novel possibility for extracting a small subset of a data base which contains all the information necessary to solve a given classification task: using the Support Vector Algorithm to train three different types of handwritten digit classifiers, we observed that these types of classifiers construct their decision surface from strongly overlapping small (k: 4%) subsets of the data base. This finding opens up the possibiiity of compressing data bases significantly by disposing of the data which is not important for the solution of a given task. In addition, we show that the theory allows us to predict the classifier that will have the best generalization ability, based solely on performance on the training set and characteristics of the learning machines. This finding is important for cases where the amount of available data is limited.

PDF [BibTex]

PDF [BibTex]