Fast and Efficient Training of RBF Networks

Buchtala, Oliver; Hofmann, Alexander; Sick, Bernhard

doi:10.1007/3-540-44989-2_6

Cited by 15 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The k-means algorithm can be regarded as a special instance of such algorithms. In [22,23] we have demonstrated that k-means can be used in combination with conventional gradient-based training algorithms such as RPROP to improve the training of RBF.…”

Section: Generative Modeling For the Detection Of Novel Attack Typesmentioning

confidence: 98%

“…Typical training algorithms are gradientbased techniques (such as backpropagation, resilient propagation, or Quickprop), clustering techniques in combination with methods for the solution of linear least-squares problems (e.g., k-means and QR decomposition), or combinations of both. An overview of various training methods is given in [22][23][24]. When RBF are used to solve classification problems, each class is typically assigned to an ''own" output neuron and an orthogonal representation of classes is used for training.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On the versatility of radial basis function neural networks: A case study in the field of intrusion detection

Fisch

Hofmann

Sick

2010

Information Sciences

Self Cite

View full text Add to dashboard Cite

Section: Generative Modeling For the Detection Of Novel Attack Typesmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

On the versatility of radial basis function neural networks: A case study in the field of intrusion detection

Fisch

Hofmann

Sick

2010

Information Sciences

Self Cite

View full text Add to dashboard Cite

“…The other J temp − 1 new centers are then chosen with a large spread in the input space (refer to Buchtala et al [2003]). (c) Run the unsupervised step of the VI algorithm using the samples in the sliding window while keeping the parameters μ j and j of the already existing rules fixed.…”

Section: Obsoleteness Detection and Reactionmentioning

confidence: 99%

Techniques for knowledge acquisition in dynamically changing environments

Fisch

Jänicke

Kalkowski

et al. 2012

ACM Trans. Auton. Adapt. Syst.

Self Cite

View full text Add to dashboard Cite

Intelligent agents often have the same or similar tasks and sometimes they cooperate to solve a given problem. These agents typically know how to observe their local environment and how to react on certain observations, for instance, and this knowledge may be represented in form of rules. However, many environments are dynamic in the sense that from time to time novel rules are required or old rules become obsolete. In this article we propose and investigate new techniques for knowledge acquisition by novelty detection and reaction as well as obsoleteness detection and reaction that an agent may use for self-adaptation to new situations. For that purpose we consider classifiers based on probabilistic rules. Premises of new rules are learned autonomously while conclusions are either obtained from human experts or from other agents which have learned appropriate rules in the past. By means of knowledge exchange, agents will efficiently be enabled to cope with situations they were not confronted with before. This kind of collaborative intelligence follows the human archetype: Humans are able to learn from each other by communicating learned rules. We demonstrate some properties of the knowledge acquisition techniques using artificial data as well as data from the field of intrusion detection.

show abstract

“…The parameters of a fuzzy classifier can be determined by means of training algorithms such as gradient-based techniques or clustering techniques in combination with methods for the solution of linear least-squares (LLS) problems (see, e.g., [11] and our own work in [12], [13]). However, linguistic terms and rules could be defined by experts as well.…”

Section: Fuzzy Classifier Paradigmmentioning

confidence: 99%

Goodness of Fit: Measures for a Fuzzy Classifier

Buchtala

Sick

2007

2007 IEEE Symposium on Foundations of Computational Intelligence

Self Cite

View full text Add to dashboard Cite

The understandability of rule sets is an important issue in knowledge discovery, where classification rules, for example, are extracted from large data sets. An important criterion in this context is the goodness of fit of a given classifier, i.e., a measure that gives an quantitative answer to the question, how good a classifier fits to the data it has to classify. In this article we provide an appropriate measure for a Mamdani-type fuzzy classifier with Gaussians and singletons as membership functions, sum-prod inference, and height method for defuzzification. That is, goodness of fit must be measured for multivariate Gaussian mixture models. Therefore, we adopt conventional test methods for univariate, unimodal probability distributions (e.g., Kolmogorov-Smirnov for chi-square), provide a measure for the goodness of fit of our fuzzy classifier, and discuss its properties. In a second step we go even beyond this point by showing how this measure could be extended to an analysis tool that gives detailed hints which rules or which membership functions are not suitably realized.

show abstract

Fast and Efficient Training of RBF Networks

Cited by 15 publications

References 14 publications

On the versatility of radial basis function neural networks: A case study in the field of intrusion detection

On the versatility of radial basis function neural networks: A case study in the field of intrusion detection

Techniques for knowledge acquisition in dynamically changing environments

Goodness of Fit: Measures for a Fuzzy Classifier

Contact Info

Product

Resources

About