@article{BornheimGriegerBlanecketal.2024, author = {Bornheim, Tobias and Grieger, Niklas and Blaneck, Patrick Gustav and Bialonski, Stephan}, title = {Speaker Attribution in German Parliamentary Debates with QLoRA-adapted Large Language Models}, series = {Journal for language technology and computational linguistics : JLCL}, volume = {37}, journal = {Journal for language technology and computational linguistics : JLCL}, number = {1}, publisher = {Gesellschaft f{\"u}r Sprachtechnologie und Computerlinguistik}, address = {Regensburg}, issn = {2190-6858}, doi = {10.21248/jlcl.37.2024.244}, pages = {13 Seiten}, year = {2024}, abstract = {The growing body of political texts opens up new opportunities for rich insights into political dynamics and ideologies but also increases the workload for manual analysis. Automated speaker attribution, which detects who said what to whom in a speech event and is closely related to semantic role labeling, is an important processing step for computational text analysis. We study the potential of the large language model family Llama 2 to automate speaker attribution in German parliamentary debates from 2017-2021. We fine-tune Llama 2 with QLoRA, an efficient training strategy, and observe our approach to achieve competitive performance in the GermEval 2023 Shared Task On Speaker Attribution in German News Articles and Parliamentary Debates. Our results shed light on the capabilities of large language models in automating speaker attribution, revealing a promising avenue for computational analysis of political discourse and the development of semantic role labeling systems.}, language = {en} } @unpublished{GriegerMehrkanoonBialonski2024, author = {Grieger, Niklas and Mehrkanoon, Siamak and Bialonski, Stephan}, title = {Preprint: Data-efficient sleep staging with synthetic time series pretraining}, series = {arXiv}, journal = {arXiv}, pages = {10 Seiten}, year = {2024}, abstract = {Analyzing electroencephalographic (EEG) time series can be challenging, especially with deep neural networks, due to the large variability among human subjects and often small datasets. To address these challenges, various strategies, such as self-supervised learning, have been suggested, but they typically rely on extensive empirical datasets. Inspired by recent advances in computer vision, we propose a pretraining task termed "frequency pretraining" to pretrain a neural network for sleep staging by predicting the frequency content of randomly generated synthetic time series. Our experiments demonstrate that our method surpasses fully supervised learning in scenarios with limited data and few subjects, and matches its performance in regimes with many subjects. Furthermore, our results underline the relevance of frequency information for sleep stage scoring, while also demonstrating that deep neural networks utilize information beyond frequencies to enhance sleep staging performance, which is consistent with previous research. We anticipate that our approach will be advantageous across a broad spectrum of applications where EEG data is limited or derived from a small number of subjects, including the domain of brain-computer interfaces.}, language = {en} } @article{RingersBialonskiEgeetal.2023, author = {Ringers, Christa and Bialonski, Stephan and Ege, Mert and Solovev, Anton and Hansen, Jan Niklas and Jeong, Inyoung and Friedrich, Benjamin M. and Jurisch-Yaksi, Nathalie}, title = {Novel analytical tools reveal that local synchronization of cilia coincides with tissue-scale metachronal waves in zebrafish multiciliated epithelia}, series = {eLife}, volume = {12}, journal = {eLife}, publisher = {eLife Sciences Publications}, issn = {2050-084X}, doi = {10.7554/eLife.77701}, pages = {27 Seiten}, year = {2023}, abstract = {Motile cilia are hair-like cell extensions that beat periodically to generate fluid flow along various epithelial tissues within the body. In dense multiciliated carpets, cilia were shown to exhibit a remarkable coordination of their beat in the form of traveling metachronal waves, a phenomenon which supposedly enhances fluid transport. Yet, how cilia coordinate their regular beat in multiciliated epithelia to move fluids remains insufficiently understood, particularly due to lack of rigorous quantification. We combine experiments, novel analysis tools, and theory to address this knowledge gap. To investigate collective dynamics of cilia, we studied zebrafish multiciliated epithelia in the nose and the brain. We focused mainly on the zebrafish nose, due to its conserved properties with other ciliated tissues and its superior accessibility for non-invasive imaging. We revealed that cilia are synchronized only locally and that the size of local synchronization domains increases with the viscosity of the surrounding medium. Even though synchronization is local only, we observed global patterns of traveling metachronal waves across the zebrafish multiciliated epithelium. Intriguingly, these global wave direction patterns are conserved across individual fish, but different for left and right noses, unveiling a chiral asymmetry of metachronal coordination. To understand the implications of synchronization for fluid pumping, we used a computational model of a regular array of cilia. We found that local metachronal synchronization prevents steric collisions, i.e., cilia colliding with each other, and improves fluid pumping in dense cilia carpets, but hardly affects the direction of fluid flow. In conclusion, we show that local synchronization together with tissue-scale cilia alignment coincide and generate metachronal wave patterns in multiciliated epithelia, which enhance their physiological function of fluid pumping.}, language = {en} } @unpublished{BornheimNiklasBlanecketal.2023, author = {Bornheim, Tobias and Niklas, Grieger and Blaneck, Patrick Gustav and Bialonski, Stephan}, title = {Preprint: Speaker attribution in German parliamentary debates with QLoRA-adapted large language models}, series = {Journal for Language Technology and Computational Linguistics}, journal = {Journal for Language Technology and Computational Linguistics}, doi = {10.48550/arXiv.2309.09902}, pages = {8 Seiten}, year = {2023}, abstract = {The growing body of political texts opens up new opportunities for rich insights into political dynamics and ideologies but also increases the workload for manual analysis. Automated speaker attribution, which detects who said what to whom in a speech event and is closely related to semantic role labeling, is an important processing step for computational text analysis. We study the potential of the large language model family Llama 2 to automate speaker attribution in German parliamentary debates from 2017-2021. We fine-tune Llama 2 with QLoRA, an efficient training strategy, and observe our approach to achieve competitive performance in the GermEval 2023 Shared Task On Speaker Attribution in German News Articles and Parliamentary Debates. Our results shed light on the capabilities of large language models in automating speaker attribution, revealing a promising avenue for computational analysis of political discourse and the development of semantic role labeling systems.}, language = {en} } @inproceedings{BlaneckBornheimGriegeretal.2022, author = {Blaneck, Patrick Gustav and Bornheim, Tobias and Grieger, Niklas and Bialonski, Stephan}, title = {Automatic readability assessment of german sentences with transformer ensembles}, series = {Proceedings of the GermEval 2022 Workshop on Text Complexity Assessment of German Text}, booktitle = {Proceedings of the GermEval 2022 Workshop on Text Complexity Assessment of German Text}, publisher = {Association for Computational Linguistics}, address = {Potsdam}, doi = {10.48550/arXiv.2209.04299}, pages = {57 -- 62}, year = {2022}, abstract = {Reliable methods for automatic readability assessment have the potential to impact a variety of fields, ranging from machine translation to self-informed learning. Recently, large language models for the German language (such as GBERT and GPT-2-Wechsel) have become available, allowing to develop Deep Learning based approaches that promise to further improve automatic readability assessment. In this contribution, we studied the ability of ensembles of fine-tuned GBERT and GPT-2-Wechsel models to reliably predict the readability of German sentences. We combined these models with linguistic features and investigated the dependence of prediction performance on ensemble size and composition. Mixed ensembles of GBERT and GPT-2-Wechsel performed better than ensembles of the same size consisting of only GBERT or GPT-2-Wechsel models. Our models were evaluated in the GermEval 2022 Shared Task on Text Complexity Assessment on data of German sentences. On out-of-sample data, our best ensemble achieved a root mean squared error of 0:435.}, language = {en} } @article{KaulenSchwabedalSchneideretal.2022, author = {Kaulen, Lars and Schwabedal, Justus T. C. and Schneider, Jules and Ritter, Philipp and Bialonski, Stephan}, title = {Advanced sleep spindle identification with neural networks}, series = {Scientific Reports}, volume = {12}, journal = {Scientific Reports}, number = {Article number: 7686}, publisher = {Springer Nature}, address = {London}, issn = {2045-2322}, doi = {10.1038/s41598-022-11210-y}, pages = {1 -- 10}, year = {2022}, abstract = {Sleep spindles are neurophysiological phenomena that appear to be linked to memory formation and other functions of the central nervous system, and that can be observed in electroencephalographic recordings (EEG) during sleep. Manually identified spindle annotations in EEG recordings suffer from substantial intra- and inter-rater variability, even if raters have been highly trained, which reduces the reliability of spindle measures as a research and diagnostic tool. The Massive Online Data Annotation (MODA) project has recently addressed this problem by forming a consensus from multiple such rating experts, thus providing a corpus of spindle annotations of enhanced quality. Based on this dataset, we present a U-Net-type deep neural network model to automatically detect sleep spindles. Our model's performance exceeds that of the state-of-the-art detector and of most experts in the MODA dataset. We observed improved detection accuracy in subjects of all ages, including older individuals whose spindles are particularly challenging to detect reliably. Our results underline the potential of automated methods to do repetitive cumbersome tasks with super-human performance.}, language = {en} } @article{RoethenbacherCesariDoppleretal.2022, author = {R{\"o}thenbacher, Annika and Cesari, Matteo and Doppler, Christopher E.J. and Okkels, Niels and Willemsen, Nele and Sembowski, Nora and Seger, Aline and Lindner, Marie and Brune, Corinna and Stefani, Ambra and H{\"o}gl, Birgit and Bialonski, Stephan and Borghammer, Per and Fink, Gereon R. and Schober, Martin and Sommerauer, Michael}, title = {RBDtector: an open-source software to detect REM sleep without atonia according to visual scoring criteria}, series = {Scientific Reports}, volume = {12}, journal = {Scientific Reports}, number = {Article number: 20886}, publisher = {Springer Nature}, address = {London}, issn = {2045-2322}, doi = {10.1038/s41598-022-25163-9}, pages = {1 -- 14}, year = {2022}, abstract = {REM sleep without atonia (RSWA) is a key feature for the diagnosis of rapid eye movement (REM) sleep behaviour disorder (RBD). We introduce RBDtector, a novel open-source software to score RSWA according to established SINBAR visual scoring criteria. We assessed muscle activity of the mentalis, flexor digitorum superficialis (FDS), and anterior tibialis (AT) muscles. RSWA was scored manually as tonic, phasic, and any activity by human scorers as well as using RBDtector in 20 subjects. Subsequently, 174 subjects (72 without RBD and 102 with RBD) were analysed with RBDtector to show the algorithm's applicability. We additionally compared RBDtector estimates to a previously published dataset. RBDtector showed robust conformity with human scorings. The highest congruency was achieved for phasic and any activity of the FDS. Combining mentalis any and FDS any, RBDtector identified RBD subjects with 100\% specificity and 96\% sensitivity applying a cut-off of 20.6\%. Comparable performance was obtained without manual artefact removal. RBD subjects also showed muscle bouts of higher amplitude and longer duration. RBDtector provides estimates of tonic, phasic, and any activity comparable to human scorings. RBDtector, which is freely available, can help identify RBD subjects and provides reliable RSWA metrics.}, language = {en} } @article{GriegerSchwabedalWendeletal.2021, author = {Grieger, Niklas and Schwabedal, Justus T. C. and Wendel, Stefanie and Ritze, Yvonne and Bialonski, Stephan}, title = {Automated scoring of pre-REM sleep in mice with deep learning}, series = {Scientific Reports}, volume = {11}, journal = {Scientific Reports}, number = {Art. 12245}, publisher = {Springer Nature}, address = {London}, issn = {2045-2322}, doi = {10.1038/s41598-021-91286-0}, year = {2021}, abstract = {Reliable automation of the labor-intensive manual task of scoring animal sleep can facilitate the analysis of long-term sleep studies. In recent years, deep-learning-based systems, which learn optimal features from the data, increased scoring accuracies for the classical sleep stages of Wake, REM, and Non-REM. Meanwhile, it has been recognized that the statistics of transitional stages such as pre-REM, found between Non-REM and REM, may hold additional insight into the physiology of sleep and are now under vivid investigation. We propose a classification system based on a simple neural network architecture that scores the classical stages as well as pre-REM sleep in mice. When restricted to the classical stages, the optimized network showed state-of-the-art classification performance with an out-of-sample F1 score of 0.95 in male C57BL/6J mice. When unrestricted, the network showed lower F1 scores on pre-REM (0.5) compared to the classical stages. The result is comparable to previous attempts to score transitional stages in other species such as transition sleep in rats or N1 sleep in humans. Nevertheless, we observed that the sequence of predictions including pre-REM typically transitioned from Non-REM to REM reflecting sleep dynamics observed by human scorers. Our findings provide further evidence for the difficulty of scoring transitional sleep stages, likely because such stages of sleep are under-represented in typical data sets or show large inter-scorer variability. We further provide our source code and an online platform to run predictions with our trained network.}, language = {en} } @inproceedings{BornheimGriegerBialonski2021, author = {Bornheim, Tobias and Grieger, Niklas and Bialonski, Stephan}, title = {FHAC at GermEval 2021: Identifying German toxic, engaging, and fact-claiming comments with ensemble learning}, series = {Proceedings of the GermEval 2021 Workshop on the Identification of Toxic, Engaging, and Fact-Claiming Comments : 17th Conference on Natural Language Processing KONVENS 2021}, booktitle = {Proceedings of the GermEval 2021 Workshop on the Identification of Toxic, Engaging, and Fact-Claiming Comments : 17th Conference on Natural Language Processing KONVENS 2021}, publisher = {Heinrich Heine University}, address = {D{\"u}sseldorf}, doi = {10.48415/2021/fhw5-x128}, pages = {105 -- 111}, year = {2021}, language = {en} } @unpublished{RingersBialonskiSolovevetal.2021, author = {Ringers, Christa and Bialonski, Stephan and Solovev, Anton and Hansen, Jan N. and Ege, Mert and Friedrich, Benjamin M. and Jurisch-Yaksi, Nathalie}, title = {Preprint: Local synchronization of cilia and tissue-scale cilia alignment are sufficient for global metachronal waves}, series = {bioRxiv}, journal = {bioRxiv}, doi = {10.1101/2021.11.23.469646}, pages = {19 Seiten}, year = {2021}, abstract = {Motile cilia are hair-like cell extensions present in multiple organs of the body. How cilia coordinate their regular beat in multiciliated epithelia to move fluids remains insufficiently understood, particularly due to lack of rigorous quantification. We combine here experiments, novel analysis tools, and theory to address this knowledge gap. We investigate collective dynamics of cilia in the zebrafish nose, due to its conserved properties with other ciliated tissues and its superior accessibility for non-invasive imaging. We revealed that cilia are synchronized only locally and that the size of local synchronization domains increases with the viscosity of the surrounding medium. Despite the fact that synchronization is local only, we observed global patterns of traveling metachronal waves across the multiciliated epithelium. Intriguingly, these global wave direction patterns are conserved across individual fish, but different for left and right nose, unveiling a chiral asymmetry of metachronal coordination. To understand the implications of synchronization for fluid pumping, we used a computational model of a regular array of cilia. We found that local metachronal synchronization prevents steric collisions and improves fluid pumping in dense cilia carpets, but hardly affects the direction of fluid flow. In conclusion, we show that local synchronization together with tissue-scale cilia alignment are sufficient to generate metachronal wave patterns in multiciliated epithelia, which enhance their physiological function of fluid pumping.}, language = {en} } @article{SchwabedalSippelBrandtetal.2018, author = {Schwabedal, Justus T. C. and Sippel, Daniel and Brandt, Moritz D. and Bialonski, Stephan}, title = {Automated Classification of Sleep Stages and EEG Artifacts in Mice with Deep Learning}, doi = {10.48550/arXiv.1809.08443}, year = {2018}, abstract = {Sleep scoring is a necessary and time-consuming task in sleep studies. In animal models (such as mice) or in humans, automating this tedious process promises to facilitate long-term studies and to promote sleep biology as a data-driven f ield. We introduce a deep neural network model that is able to predict different states of consciousness (Wake, Non-REM, REM) in mice from EEG and EMG recordings with excellent scoring results for out-of-sample data. Predictions are made on epochs of 4 seconds length, and epochs are classified as artifactfree or not. The model architecture draws on recent advances in deep learning and in convolutional neural networks research. In contrast to previous approaches towards automated sleep scoring, our model does not rely on manually defined features of the data but learns predictive features automatically. We expect deep learning models like ours to become widely applied in different fields, automating many repetitive cognitive tasks that were previously difficult to tackle.}, language = {en} } @article{KarnatakKantzBialonski2017, author = {Karnatak, Rajat and Kantz, Holger and Bialonski, Stephan}, title = {Early warning signal for interior crises in excitable systems}, series = {Physical Review E}, volume = {96}, journal = {Physical Review E}, number = {4}, issn = {2470-0053}, doi = {10.1103/PhysRevE.96.042211}, pages = {042211}, year = {2017}, language = {en} } @article{NgamgaBialonskiMarwanetal.2016, author = {Ngamga, Eulalie Joelle and Bialonski, Stephan and Marwan, Norbert and Kurths, J{\"u}rgen and Geier, Christian and Lehnertz, Klaus}, title = {Evaluation of selected recurrence measures in discriminating pre-ictal and inter-ictal periods from epileptic EEG data}, series = {Physics Letters A}, volume = {380}, journal = {Physics Letters A}, number = {16}, publisher = {Elsevier}, address = {Amsterdam}, issn = {0375-9601}, doi = {10.1016/j.physleta.2016.02.024}, pages = {1419 -- 1425}, year = {2016}, abstract = {We investigate the suitability of selected measures of complexity based on recurrence quantification analysis and recurrence networks for an identification of pre-seizure states in multi-day, multi-channel, invasive electroencephalographic recordings from five epilepsy patients. We employ several statistical techniques to avoid spurious findings due to various influencing factors and due to multiple comparisons and observe precursory structures in three patients. Our findings indicate a high congruence among measures in identifying seizure precursors and emphasize the current notion of seizure generation in large-scale epileptic networks. A final judgment of the suitability for field studies, however, requires evaluation on a larger database.}, language = {en} } @article{BialonskiCaronSchloenetal.2016, author = {Bialonski, Stephan and Caron, David A. and Schloen, Julia and Feudel, Ulrike and Kantz, Holger and Moorthi, Stefanie D.}, title = {Phytoplankton dynamics in the Southern California Bight indicate a complex mixture of transport and biology}, series = {Journal of Plankton Research}, volume = {38}, journal = {Journal of Plankton Research}, number = {4}, publisher = {Oxford University Press}, address = {Oxford}, issn = {1464-3774}, doi = {10.1093/plankt/fbv122}, pages = {1077 -- 1091}, year = {2016}, abstract = {The stimulation and dominance of potentially harmful phytoplankton taxa at a given locale and time are determined by local environmental conditions as well as by transport to or from neighboring regions. The present study investigated the occurrence of common harmful algal bloom (HAB) taxa within the Southern California Bight, using cross-correlation functions to determine potential dependencies between HAB taxa and environmental factors, and potential links to algal transport via local hydrography and currents. A simulation study, in which Lagrangian particles were released, was used to assess travel times due to advection by prevailing ocean currents in the bight. Our results indicate that transport of some taxa may be an important mechanism for the expansion of their distributions into other regions, which was supported by mean travel times derived from our simulation study and other literature on ocean currents in the Southern California Bight. In other cases, however, phytoplankton dynamics were rather linked to local environmental conditions, including coastal upwelling events. Overall, our study shows that complex current patterns in the Southern California Bight may contribute significantly to the formation and expansion of HABs in addition to local environmental factors determining the spatiotemporal dynamics of phytoplankton blooms.}, language = {en} } @incollection{Bialonski2016, author = {Bialonski, Stephan}, title = {Are interaction clusters in epileptic networks predictive of seizures?}, series = {Epilepsy: The Intersection of Neurosciences, Biology, Mathematics, Engineering, and Physics}, booktitle = {Epilepsy: The Intersection of Neurosciences, Biology, Mathematics, Engineering, and Physics}, publisher = {CRC Press}, isbn = {978-143983886-0}, pages = {349 -- 355}, year = {2016}, language = {en} } @article{BialonskiAnsmannKantz2015, author = {Bialonski, Stephan and Ansmann, Gerrit and Kantz, Holger}, title = {Data-driven prediction and prevention of extreme events in a spatially extended excitable system}, series = {Physical Review E}, volume = {92}, journal = {Physical Review E}, number = {4}, issn = {2470-0053}, doi = {10.1103/PhysRevE.92.042910}, pages = {042910}, year = {2015}, language = {en} } @article{MulhernBialonskiKantz2015, author = {Mulhern, Colm and Bialonski, Stephan and Kantz, Holger}, title = {Extreme events due to localization of energy}, series = {Physical Review E}, volume = {91}, journal = {Physical Review E}, number = {1}, issn = {2470-0053}, doi = {10.1103/PhysRevE.91.012918}, pages = {012918}, year = {2015}, language = {en} } @article{GeierLehnertzBialonski2015, author = {Geier, Christian and Lehnertz, Klaus and Bialonski, Stephan}, title = {Time-dependent degree-degree correlations in epileptic brain networks: from assortative to dissortative mixing}, series = {Frontiers in Human Neuroscience}, journal = {Frontiers in Human Neuroscience}, publisher = {Frontiers Research Foundation}, address = {Lausanne}, issn = {1662-5161}, doi = {10.3389/fnhum.2015.00462}, year = {2015}, language = {en} } @article{GeierBialonskiElgeretal.2015, author = {Geier, Christian and Bialonski, Stephan and Elger, Christian E. and Lehnertz, Klaus}, title = {How important is the seizure onset zone for seizure dynamics?}, series = {Seizure}, volume = {25}, journal = {Seizure}, issn = {1059-1311}, doi = {10.1016/j.seizure.2014.10.013}, pages = {160 -- 166}, year = {2015}, language = {en} } @article{LehnertzAnsmannBialonskietal.2014, author = {Lehnertz, Klaus and Ansmann, Gerrit and Bialonski, Stephan and Dickten, Henning and Geier, Christian and Porz, Stephan}, title = {Evolving networks in the human epileptic brain}, series = {Physica D: Nonlinear Phenomena}, volume = {267}, journal = {Physica D: Nonlinear Phenomena}, publisher = {Elsevier}, address = {Amsterdam}, issn = {0167-2789}, doi = {10.1016/j.physd.2013.06.009}, pages = {7 -- 15}, year = {2014}, abstract = {Network theory provides novel concepts that promise an improved characterization of interacting dynamical systems. Within this framework, evolving networks can be considered as being composed of nodes, representing systems, and of time-varying edges, representing interactions between these systems. This approach is highly attractive to further our understanding of the physiological and pathophysiological dynamics in human brain networks. Indeed, there is growing evidence that the epileptic process can be regarded as a large-scale network phenomenon. We here review methodologies for inferring networks from empirical time series and for a characterization of these evolving networks. We summarize recent findings derived from studies that investigate human epileptic brain networks evolving on timescales ranging from few seconds to weeks. We point to possible pitfalls and open issues, and discuss future perspectives.}, language = {en} }