K
Klaus-Robert Müller
Researcher at Technical University of Berlin
Publications - 799
Citations - 98394
Klaus-Robert Müller is an academic researcher from Technical University of Berlin. The author has contributed to research in topics: Artificial neural network & Computer science. The author has an hindex of 129, co-authored 764 publications receiving 79391 citations. Previous affiliations of Klaus-Robert Müller include Korea University & University of Tokyo.
Papers
More filters
Proceedings Article
Inducing Metric Violations in Human Similarity Judgements
TL;DR: There may not be a strict dichotomy between either a metric or a non-metric internal space but rather degrees to which potentially large subsets of stimuli are represented metrically with a small subset causing a global violation of metricity.
Proceedings Article
XAI for Transformers: Better Explanations through Conservative Propagation
TL;DR: This proposal, which can be seen as a proper extension of the well-established LRP method to Transformers, is shown both theoretically and empirically to overcome the deficiency of a simple gradient-based approach, and achieves state-of-the-art explanation performance on a broad range of Transformer models and datasets.
Proceedings ArticleDOI
Covariate shift adaptation in EMG pattern recognition for prosthetic device control.
Marina M.-C. Vidovic,Liliana P. Paredes,Hwang Han-Jeong Hwang,Sebastian Amsüss,Jaspar Pahl,Janne M. Hahne,Bernhard Graimann,Dario Farina,Klaus-Robert Müller +8 more
TL;DR: The results showed that an estimator that shrinks the training model parameters towards the calibration set parameters significantly increased the classifier performance across different testing days, indicating that the proposed methodology can be a practical means for improving robustness in pattern recognition methods for myocontrol.
Posted Content
Counterstrike: Defending Deep Learning Architectures Against Adversarial Samples by Langevin Dynamics with Supervised Denoising Autoencoder.
TL;DR: A novel effective strategy that allows to relax adversarial samples onto the underlying manifold of the (unknown) target class distribution and exhibits a high robustness against blackbox and whitebox attacks and outperforms state-of-the-art defense algorithms.
Journal ArticleDOI
Canonical maximization of coherence: A novel tool for investigation of neuronal interactions between two datasets
Carmen Vidaurre,Carmen Vidaurre,Guido Nolte,I. E. J. de Vries,Marisol Gómez,Tjeerd W. Boonstra,Tjeerd W. Boonstra,Klaus-Robert Müller,Arno Villringer,Arno Villringer,Vadim V. Nikulin,Vadim V. Nikulin,Vadim V. Nikulin +12 more
TL;DR: The caCOH algorithm is presented, which maximizes the absolute value of the coherence between the two multivariate spaces in the frequency domain and allows very fast optimization for many frequency bins.