FG Maschinelles Lernen

58 Items

Recent Submissions
Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology

Studer, Stefan ; Bui, Thanh Binh ; Drescher, Christian ; Hanuschkin, Alexander ; Winkler, Ludwig ; Peters, Steven ; Müller, Klaus-Robert (2021-04-22)

Machine learning is an established and frequently used technique in industry and academia, but a standard process model to improve success and efficiency of machine learning applications is still missing. Project organizations and machine learning practitioners face manifold challenges and risks when developing machine learning applications and have a need for guidance to meet business expectat...

Identification of nodes and Networks

Liu, Yang (2021)

Complex systems in a broad range of scientific domains have been shown to be well-characterized by networks in an increasing number of studies. Problems such as cascading failures, spreading dynamics and the extraction of leading factors from raw data through the construction of networks can all be studied within the paradigm of network science. Such problems concerning networks are usually dir...

Dynamical strengthening of covalent and non-covalent molecular interactions by nuclear quantum effects at finite temperature

Sauceda, Huziel E. ; Vassilev-Galindo, Valentin ; Chmiela, Stefan ; Müller, Klaus-Robert ; Tkatchenko, Alexandre (2021-01-19)

Nuclear quantum effects (NQE) tend to generate delocalized molecular dynamics due to the inclusion of the zero point energy and its coupling with the anharmonicities in interatomic interactions. Here, we present evidence that NQE often enhance electronic interactions and, in turn, can result in dynamical molecular stabilization at finite temperature. The underlying physical mechanism promoted b...

Sensorimotor functional connectivity: A neurophysiological factor related to BCI performance

Vidaurre, Carmen ; Haufe, Stefan ; Jorajuría, Tania ; Müller, Klaus-Robert ; Nikulin, Vadim V. (2020-12-18)

Brain-Computer Interfaces (BCIs) are systems that allow users to control devices using brain activity alone. However, the ability of participants to command BCIs varies from subject to subject. About 20% of potential users of sensorimotor BCIs do not gain reliable control of the system. The inefficiency to decode user's intentions requires the identification of neurophysiological factors determ...

Machine learning methods for modeling gaze allocation in simple choice behavior and functional neuroimaging data on the level of the individual

Thomas, Armin W. (2020)

The work presented in this thesis uses tools from machine learning, statistics, and computation to build analysis methods for capturing individual differences in two research questions from cognitive neuroscience. The first chapter of this thesis investigates the role of looking behavior in simple choices of individuals (e.g., deciding whether to eat an apple or a banana for breakfast). To th...

Change-point detection in wind turbine SCADA data for robust condition monitoring with normal behaviour models

Letzgus, Simon (2020-10-27)

Analysis of data from wind turbine supervisory control and data acquisition (SCADA) systems has attracted considerable research interest in recent years. Its predominant application is to monitor turbine condition without the need for additional sensing equipment. Most approaches apply semi-supervised anomaly detection methods, also called normal behaviour models, that require clean training da...

ML2Motif—Reliable extraction of discriminative sequence motifs from learning machines

Vidovic, Marina Marie-Claire ; Kloft, Marius ; Müller, Klaus-Robert ; Görnitz, Nico (2017-03-27)

High prediction accuracies are not the only objective to consider when solving problems using machine learning. Instead, particular scientific applications require some explanation of the learned prediction function. For computational biology, positional oligomer importance matrices (POIMs) have been successfully applied to explain the decision of support vector machines (SVMs) using weighted-d...

A convolutional neural network for steady state visual evoked potential classification under ambulatory environment

Kwak, No-Sang ; Müller, Klaus-Robert ; Lee, Seong-Whan (2017-02-22)

The robust analysis of neural signals is a challenging problem. Here, we contribute a convolutional neural network (CNN) for the robust classification of a steady-state visual evoked potentials (SSVEPs) paradigm. We measure electroencephalogram (EEG)-based SSVEPs for a brain-controlled exoskeleton under ambulatory conditions in which numerous artifacts may deteriorate decoding. The proposed CNN...

Molecular force fields with gradient-domain machine learning (GDML): Comparison and synergies with classical force fields

Sauceda, Huziel E. ; Gastegger, Michael ; Chmiela, Stefan ; Müller, Klaus-Robert ; Tkatchenko, Alexandre (2020-09-24)

Modern machine learning force fields (ML-FF) are able to yield energy and force predictions at the accuracy of high-level ab initio methods, but at a much lower computational cost. On the other hand, classical molecular mechanics force fields (MM-FF) employ fixed functional forms and tend to be less accurate, but considerably faster and transferable between molecules of the same class. In this ...

Beyond Pairwise Interactions: The Totally Antisymmetric Part of the Bispectrum as Coupling Measure of at Least Three Interacting Sources

Bartz, Sarah ; Andreou, Christina ; Nolte, Guido (2020-10-26)

In this paper we make two contributions to the analysis of brain oscillations with CFC techniques. First, we introduce a new bispectral CFC measure which is selective to couplings between three or more brain sources. This measure can be derived from ordinary cross-bispectra by performing a total-antisymmetrization operation on them. Significant coupling values can then be attributed to at least...

Immediate brain plasticity after one hour of brain–computer interface (BCI)

Nierhaus, Till ; Vidaurre, Carmen ; Sannelli, Claudia ; Müller, Klaus-Robert ; Villringer, Arno (2019-11-06)

A brain‐computer‐interface (BCI) allows humans to control computational devices using only neural signals. However, it is still an open question, whether performing BCI also impacts on the brain itself, i.e. whether brain plasticity is induced. Here, we show rapid and spatially specific signs of brain plasticity measured with functional and structural MRI after only 1 h of purely mental BCI tra...

Development of model observers for quantitative assessment of mammography image quality

Kretz, Tobias (2020)

Assurance of image quality has become a basic need in our society as images play a crucial role in this era of social media and digitization. Applications range from surveillance to medical imaging. Image quality is defined as the degree of clarity of its elements. Images undergo a chain of processes before being perceived by the viewer, starting from acquisition to digitization, compression, p...

Exploring density functional subspaces with genetic algorithms

Gastegger, Michael ; González, Leticia ; Marquetand, Philipp (2018-12-14)

We use a genetic algorithm to explore the subspace of combination and parametrization patterns spanned by a set of popular exchange and correlation functional approximations. Using the well-balanced GMTKN30 benchmark database to guide the evolutionary process, we find that the genetic algorithm is able to recover variants of several popular generalized gradient approximation functionals and hyb...

Combining SchNet and SHARC: The SchNarc Machine Learning Approach for Excited-State Dynamics

Westermayr, Julia ; Gastegger, Michael ; Marquetand, Philipp (2020-04-20)

In recent years, deep learning has become a part of our everyday life and is revolutionizing quantum chemistry as well. In this work, we show how deep learning can be used to advance the research field of photochemistry by learning all important properties—multiple energies, forces, and different couplings—for photodynamics simulations. We simplify such simulations substantially by (i) a phase-...

Machine learning enables long time scale molecular photodynamics simulations

Westermayr, Julia ; Gastegger, Michael ; Menger, Maximilian F. S. J. ; Mai, Sebastian ; González, Leticia ; Marquetand, Philipp (2019-08-05)

Photo-induced processes are fundamental in nature but accurate simulations of their dynamics are seriously limited by the cost of the underlying quantum chemical calculations, hampering their application for long time scales. Here we introduce a method based on machine learning to overcome this bottleneck and enable accurate photodynamics on nanosecond time scales, which are otherwise out of re...

Times Are Changing: Investigating the Pace of Language Change in Diachronic Word Embeddings

Brandl, Stephanie ; Lassner, David (2019-08-02)

We propose Word Embedding Networks, a novel method that is able to learn word embeddings of individual data slices while simultaneously aligning and ordering them without feeding temporal information a priori to the model. This gives us the opportunity to analyse the dynamics in word embeddings on a large scale in a purely data-driven manner. In experiments on two different newspaper corpora, t...

Quantum-Chemical Insights from Interpretable Atomistic Neural Networks

Schütt, Kristof T. ; Gastegger, Michael ; Tkatchenko, Alexandre ; Müller, Klaus-Robert (2019-09-10)

With the rise of deep neural networks for quantum chemistry applications, there is a pressing need for architectures that, beyond delivering accurate predictions of chemical properties, are readily interpretable by researchers. Here, we describe interpretation techniques for atomistic neural networks on the example of Behler–Parrinello networks as well as the end-to-end model SchNet. Both model...

Molecular Dynamics with Neural Network Potentials

Gastegger, Michael ; Marquetand, Philipp (2020-06-04)

Molecular dynamics simulations are an important tool for describing the evolution of a chemical system with time. However, these simulations are inherently held back either by the prohibitive cost of accurate electronic structure theory computations or the limited accuracy of classical empirical force fields. Machine learning techniques can help to overcome these limitations by providing access...

Hamiltonian datasets: "Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions""

Schütt, Kristof T. ; Gastegger, Michael ; Tkatchenko, Alexandre ; Müller, Klaus-Robert ; Maurer, Reinhard J. (2019-09-25)

Machine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analys...

Similarity encoder: A neural network architecture for learning similarity preserving embeddings

Horn, Franziska (2020)

Matrix factorization is at the heart of many machine learning algorithms, for example, for dimensionality reduction (e.g. kernel PCA) or recommender systems relying on collaborative filtering. Understanding a singular value decomposition (SVD) of a matrix as a neural network optimization problem enables us to decompose large matrices efficiently while dealing naturally with missing values in th...