Activation Functions for Generalized Learning Vector Quantization - A Performance Comparison

Autor: Villmann, Thomas, Ravichandran, John, Villmann, Andrea, Nebel, David, Kaden, Marika

An appropriate choice of the activation function (like ReLU, sigmoid or swish) plays an important role in the performance of (deep) multilayer perceptrons (MLP) for classification and regression learning. Prototype-based classification learning metho

Externí odkaz: http://arxiv.org/abs/1901.05995

Learning Vector Quantization with Adaptive Cost-Based Outlier-Rejection

Autor: Villmann, Thomas, Kaden, Marika, Nebel, David, Biehl, Michael, Azzopardi, George, Petkov, Nicolai

Publikováno v: Computer Analysis of Images and Patterns ISBN: 9783319231167
CAIP (2)
Computer Analysis of Images and Patterns, 772-782
STARTPAGE=772;ENDPAGE=782;TITLE=Computer Analysis of Images and Patterns

We consider a reject option for prototype-based Learning Vector Quantization LVQ, which facilitates the detection of outliers in the data during the classification process. The rejection mechanism is based on a distance-based criterion and the corres

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7eb42d15da97fbf0c7df868ef8015652
https://doi.org/10.1007/978-3-319-23117-4_66

Rejection Strategies for Learning Vector Quantization – A Comparison of Probabilistic and Deterministic Approaches

Autor: Fischer, Lydia, Nebel, David, Villmann, Thomas, Hammer, Barbara, Wersing, Heiko, Schleif, Frank-Michael, Kaden, Marika, Lange, Mandy

Publikováno v: Advances in Self-Organizing Maps and Learning Vector Quantization ISBN: 9783319076942
WSOM

In this contribution, we focus on reject options for prototype-based classifiers, and we present a comparison of reject options based on statistical models for prototype-based classification as compared to alternatives which are motivated by simple g

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9a2fd0be18fa0221b8bb64a56326d1cf
https://doi.org/10.1007/978-3-319-07695-9_10

Adaptive Hausdorff Distances and Tangent Distance Adaptation for Transformation Invariant Classification Learning.

Autor: Saralajew, Sascha, Nebel, David, Villmann, Thomas

Publikováno v: Neural Information Processing: 23rd International Conference, ICONIP 2016, Kyoto, Japan, October 16-21, 2016, Proceedings, Part III; 2016, p362-371, 10p

Partial Mutual Information for Classification of Gene Expression Data by Learning Vector Quantization.

Autor: Lange, Mandy, Nebel, David, Villmann, Thomas

Publikováno v: Advances in Self-Organizing Maps & Learning Vector Quantization; 2014, p259-269, 11p