Maximilian Münch
Research Subjects
Maximilian Münch has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Frank-Michael Schleif
Technical University of Applied Sciences Würzburg-Schweinfurt
4 shared publications Christoph Raab
2 shared publicationsFrank-Michael Schleif
University of Applied Sciences Würzburg-Schweinfurt
2 shared publications- Manuel Röder
Technical University of Applied Sciences Würzburg-Schweinfurt, Center for Artificial Intelligence and Robotics, Würzburg, Germany
2 shared publications - Katrin Sophie Bohnsack
Univ Appl Sci Mittweida, Saxon Inst Computat Intelligence & Machine Learnin, D-09648 Mittweida, Germany
1 shared publication - Michael Biehl
Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, University of Groningen, Groningen, the Netherlands; Department of Metabolism and System Science, School of Medical Sciences, College of Medicine and Health, University of Birmingham, Birmingham B152TT, UK
1 shared publication
Latest Publications
- Open Access
- 28 June 2024
- Published by Elsevier in Neurocomputing
- Vol. 596
Data-distribution-informed Nyström approximation for structured data using vector quantization-based landmark determination
- 28 June 2024
- Published by Elsevier in Neurocomputing
- Vol. 596
Open AccessData-distribution-informed Nyström approximation for structured data using vector quantization-based landmark determination
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1867-1876
Unlocking the Potential of Non-PSD Kernel Matrices: A Polar Decomposition-based Transformation for Improved Prediction Models
Kernel functions are a key element in many machine learning methods to capture the similarity between data points. However, a considerable number of these functions do not meet all mathematical requirements to be a valid positive semi-definite kernel, a crucial precondition for kernel-based classifiers such as Support Vector Machines or Kernel Fisher Discriminant classifiers. In this paper, we propose a novel strategy employing a polar decomposition to effectively transform invalid kernel matrices to positive semi-definite matrices, while preserving the topological structure inherent to the data points. Utilizing polar decomposition allows the effective transformation of indefinite kernel matrices from Krein space to positive semi-definite matrices in Hilbert space, thereby providing an efficient out-of-sample extension for new unseen data and enhancing kernel method applicability across diverse classification tasks. We evaluate our approach on a variety of benchmark datasets and demonstrate its superiority over competitive methods.
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1867-1876
Open AccessUnlocking the Potential of Non-PSD Kernel Matrices: A Polar Decomposition-based Transformation for Improved Prediction Models
Kernel functions are a key element in many machine learning methods to capture the similarity between data points. However, a considerable number of these functions do not meet all mathematical requirements to be a valid positive semi-definite kernel, a crucial precondition for kernel-based classifiers such as Support Vector Machines or Kernel Fisher Discriminant classifiers. In this paper, we propose a novel strategy employing a polar decomposition to effectively transform invalid kernel matrices to positive semi-definite matrices, while preserving the topological structure inherent to the data points. Utilizing polar decomposition allows the effective transformation of indefinite kernel matrices from Krein space to positive semi-definite matrices in Hilbert space, thereby providing an efficient out-of-sample extension for new unseen data and enhancing kernel method applicability across diverse classification tasks. We evaluate our approach on a variety of benchmark datasets and demonstrate its superiority over competitive methods.
- 28 July 2023
- Published by Elsevier in Neurocomputing
- Vol. 555
Static and adaptive subspace information fusion for indefinite heterogeneous proximity data
- 28 July 2023
- Published by Elsevier in Neurocomputing
- Vol. 555
Static and adaptive subspace information fusion for indefinite heterogeneous proximity data
- 10 April 2021
- Published by Springer Nature
- p. 14-23
Complex-Valued Embeddings of Generic Proximity Data
- 10 April 2021
- Published by Springer Nature
- p. 14-23
Complex-Valued Embeddings of Generic Proximity Data
- 23 December 2020
- Published by Springer Nature
- p. 112-137
Encoding of Indefinite Proximity Data: A Structure Preserving Perspective
- 23 December 2020
- Published by Springer Nature
- p. 112-137
Encoding of Indefinite Proximity Data: A Structure Preserving Perspective
- 16 May 2019
- Published by Springer Nature
- p. 799-810
Device-Free Passive Human Counting with Bluetooth Low Energy Beacons
- 16 May 2019
- Published by Springer Nature
- p. 799-810
Device-Free Passive Human Counting with Bluetooth Low Energy Beacons