Fuzzy classification by fuzzy labeled neural gas

Villmann, T.; Hammer, Barbara; Schleif, Frank-Michael; Geweniger, Tina; Herrmann, W.

doi:10.1016/j.neunet.2006.05.026

Cited by 30 publications

(16 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on these findings, the question arises whether further metaparameters of the learning scheme can be determined in a similar way. This includes, for example, the labeling of prototypes which has been made adaptive in [20].…”

Section: Resultsmentioning

confidence: 99%

Hyperparameter learning in probabilistic prototype-based models

2010

View full text Add to dashboard Cite

a b s t r a c tWe present two approaches to extend Robust Soft Learning Vector Quantization (RSLVQ). This algorithm for nearest prototype classification is derived from an explicit cost function and follows the dynamics of a stochastic gradient ascent. The RSLVQ cost function is defined in terms of a likelihood ratio and involves a hyperparameter which is kept constant during training. We propose to adapt the hyperparameter in the training phase based on the gradient information. Besides, we propose to base the classifier's decision on the value of the likelihood ratio instead of using the distance based classification approach. Experiments on artificial and real life data show that the hyperparameter crucially influences the performance of RSLVQ. However, it is not possible to estimate the best value from the data prior to learning. We show that the proposed variant of RSLVQ is very robust with respect to the initial value of the hyperparameter. The classification approach based on the likelihood ratio turns out to be superior to distance based classification, if local hyperparameters are adapted for each prototype.

show abstract

Section: Resultsmentioning

confidence: 99%

Hyperparameter learning in probabilistic prototype-based models

2010

View full text Add to dashboard Cite

show abstract

“…This attempt was done in [31] for FSNPC. An alternative for those problems might be a combination of unsupervised vector quantization together with an supervised fuzzy classifier extension based on self-organizing maps (SOM, [51,52]) and neural gas (NG, [60,59]) as proposed in [105].…”

Section: Fuzzy Data and Fuzzy Classification Approaches Related To Lvqmentioning

confidence: 99%

Aspects in Classification Learning - Review of Recent Developments in Learning Vector Quantization

Kaden¹,

Lange²,

Nebel³

et al. 2014

Foundations of Computing and Decision Sciences

Self Cite

View full text Add to dashboard Cite

Abstract. Classification is one of the most frequent tasks in machine learning. However, the variety of classification tasks as well as classifier methods is huge. Thus the question is coming up: which classifier is suitable for a given problem or how can we utilize a certain classifier model for different tasks in classification learning. This paper focuses on learning vector quantization classifiers as one of the most intuitive prototype based classification models. Recent extensions and modifications of the basic learning vector quantization algorithm, which are proposed in the last years, are highlighted and also discussed in relation to particular classification task scenarios like imbalanced and/or incomplete data, prior data knowledge, classification guarantees or adaptive data metrics for optimal classification.

show abstract

“…First promising steps into this direction can be found in the approach [15] for online NG, however, the method proposed in [15] cannot be transferred to median versions. A batch variant, supervised batch NG (SBNG), recently proposed in [6], is based on the cost…”

Section: Supervised Neural Gasmentioning

confidence: 99%

Supervised Median Clustering

Hammer¹,

Hasenfuss²,

Schleif³

et al. 2006

Intelligent Engineering Systems Through Artificial Neural Networks, Volume 16

View full text Add to dashboard Cite

Prototype based clustering and classification algorithms constitute very intuitive and powerful machine learning tools for a variety of application areas. They combine simple training algorithms and easy interpretability by means of prototype inspection. However, the classical methods are restricted to data embedded in a real vector space and thus, have only limited applicability to complex data as occurs in bioinformatics or symbolic areas. Recently, extensions of unsupervised prototype based clustering to proximity data, i.e. data characterized in terms of a distance matrix only, have been proposed. Since the distance matrix constitutes a universal interface, this opens the way towards an application of efficient prototype based methods for general data. In this contribution, we transfer this idea to supervised scenarios, proposing a prototype based classification method for general proximity data.

show abstract

Fuzzy classification by fuzzy labeled neural gas

Cited by 30 publications

References 14 publications

Hyperparameter learning in probabilistic prototype-based models

Hyperparameter learning in probabilistic prototype-based models

Aspects in Classification Learning - Review of Recent Developments in Learning Vector Quantization

Supervised Median Clustering

Contact Info

Product

Resources

About