Deutsches Forschungszentrum für Künstliche Intelligenz
Recent publications
Modulation of commensal gut microbiota is increasingly recognized as a promising strategy to reduce mortality in patients with malignant diseases, but monitoring for dysbiosis is generally not routine clinical practice due to equipment, expertise and funding required for sequencing analysis. A low-threshold alternative is microbial diversity profiling by single-cell flow cytometry (FCM), which we compared to 16S rRNA sequencing in human fecal samples and employed to characterize longitudinal changes in the microbiome composition of patients with aggressive B-cell non-Hodgkin lymphoma undergoing chemoimmunotherapy. Diversity measures obtained from both methods were correlated and captured identical trends in microbial community structures, finding no difference in patients' pretreatment alpha or beta diversity compared to healthy controls and a significant and progressive loss of alpha diversity during chemoimmunotherapy. Our results highlight the potential of FCM-based microbiome profiling as a reliable and accessible diagnostic tool that can provide novel insights into cancer therapy-associated dysbiosis dynamics.
Load forecasting avoids energy wastage by accurately estimating the future quantity of energy generation and demand. Existing load forecasting approaches do not utilize the potential of feature selection and dimensionality reduction approaches that remove irrelevant/redundant features and improve the performance of machine learning (ML) regressors. This research presents an end-to-end framework named Energy Generation and Demand forecasting Search Net (EGD-SNet) capable of predicting energy generation, demand and temperature in multiple regions. EGD-SNet framework contains 13 different feature selection and 11 dimensionality reduction algorithms along with 10 most widely used ML regressors. It makes use of Particle Swarm Optimizer (PSO) to smartly train regressors by finding optimal hyperparameters. Further, it has potential to design an end to end pipeline by finding appropriate combination of regressor and feature selection or dimensionality reduction approaches for precisely predicting energy generation or demand for a particular regional data based on the characteristics of data. EGD-SNet as web service is accessible here.
Background and objective Interactions of long non-coding ribonucleic acids (lncRNAs) with micro-ribonucleic acids (miRNAs) play an essential role in gene regulation, cellular metabolic, and pathological processes. Existing purely sequence based computational approaches lack robustness and efficiency mainly due to the high length variability of lncRNA sequences. Hence, the prime focus of the current study is to find optimal length trade-offs between highly flexible length lncRNA sequences. Method The paper at hand performs in-depth exploration of diverse copy padding, sequence truncation approaches, and presents a novel idea of utilizing only subregions of lncRNA sequences to generate fixed-length lncRNA sequences. Furthermore, it presents a novel bag of tricks-based deep learning approach “Bot-Net” which leverages a single layer long-short-term memory network regularized through DropConnect to capture higher order residue dependencies, pooling to retain most salient features, normalization to prevent exploding and vanishing gradient issues, learning rate decay, and dropout to regularize precise neural network for lncRNA–miRNA interaction prediction. Results BoT-Net outperforms the state-of-the-art lncRNA–miRNA interaction prediction approach by 2%, 8%, and 4% in terms of accuracy, specificity, and matthews correlation coefficient. Furthermore, a case study analysis indicates that BoT-Net also outperforms state-of-the-art lncRNA–protein interaction predictor on a benchmark dataset by accuracy of 10%, sensitivity of 19%, specificity of 6%, precision of 14%, and matthews correlation coefficient of 26%. Conclusion In the benchmark lncRNA–miRNA interaction prediction dataset, the length of the lncRNA sequence varies from 213 residues to 22,743 residues and in the benchmark lncRNA–protein interaction prediction dataset, lncRNA sequences vary from 15 residues to 1504 residues. For such highly flexible length sequences, fixed length generation using copy padding introduces a significant level of bias which makes a large number of lncRNA sequences very much identical to each other and eventually derail classifier generalizeability. Empirical evaluation reveals that within 50 residues of only the starting region of long lncRNA sequences, a highly informative distribution for lncRNA–miRNA interaction prediction is contained, a crucial finding exploited by the proposed BoT-Net approach to optimize the lncRNA fixed length generation process. Availability BoT-Net web server can be accessed at Graphic Abstract
Machine learning on trees has been mostly focused on trees as input. Much less research has investigated trees as output, which has many applications, such as molecule optimization for drug discovery, or hint generation for intelligent tutoring systems. In this work, we propose a novel autoencoder approach, called recursive tree grammar autoencoder (RTG-AE), which encodes trees via a bottom-up parser and decodes trees via a tree grammar, both learned via recursive neural networks that minimize the variational autoencoder loss. The resulting encoder and decoder can then be utilized in subsequent tasks, such as optimization and time series prediction. RTG-AEs are the first model to combine three features: recursive processing, grammatical knowledge, and deep learning. Our key message is that this unique combination of all three features outperforms models which combine any two of the three. Experimentally, we show that RTG-AE improves the autoencoding error, training time, and optimization score on synthetic as well as real datasets compared to four baselines. We further prove that RTG-AEs parse and generate trees in linear time and are expressive enough to handle all regular tree grammars.
Zusammenfassung Spätestens seit Beginn der Corona-Pandemie ist eine Vielzahl von Unternehmen aufgrund von Social-Distancing-Maßnahmen und neuen Mitarbeiteranforderungen mit hybrider Arbeit als neuem Status Quo konfrontiert. Während hierin absehbar das Potenzial zur generellen Flexibilisierung von Arbeit liegt, müssen Arbeitgeber veränderte Formen der Zusammenarbeit auch ermöglichen und räumliche Distanzen überbrücken. Hinzu kommt, dass der unerwartete Eintritt der pandemischen Lage oftmals zur ad-hoc-Einführung von Werkzeugen geführt hat, die als „Erste Hilfe“-Lösung grundsätzliche Kommunikation und Kollaboration ermöglichen sollten. Knapp drei Jahre nach Beginn der Pandemie vergleichen wir gemeinsam mit Initiatoren und Nutzern zwei derartige Anwendungsfälle, die die Metaverse-Lösung Gather eingeführt haben. In einer multiperspektivischen Untersuchung fragen wir nach den geplanten und realisierten Mehrwerten und stellen Verbindungen zu den Features der Software her. Aus den Erkenntnissen leiten wir zwei Arten von Handlungsempfehlungen ab. Einerseits, inwiefern Gather ad-hoc als „Erste Hilfe“-Lösung geeignet ist und andererseits, welche Potenziale in der zukünftigen Überarbeitung gehoben werden können, um ein robustes hybrides Arbeitsmodell aufzubauen.
Page object detection in scanned document images is a complex task due to varying document layouts and diverse page objects. In the past, traditional methods such as Optical Character Recognition (OCR)-based techniques have been employed to extract textual information. However, these methods fail to comprehend complex page objects such as tables and figures. This paper addresses the localization problem and classification of graphical objects that visually summarize vital information in documents. Furthermore, this work examines the benefit of incorporating attention mechanisms in different object detection networks to perform page object detection on scanned document images. The model is designed with a Pytorch-based framework called Detectron2. The proposed pipelines can be optimized end-to-end and exhaustively evaluated on publicly available datasets such as DocBank, PublayNet, and IIIT-AR-13K. The achieved results reflect the effectiveness of incorporating the attention mechanism for page object detection in documents.
Circular ribonucleic acids (circRNAs) are novel non-coding RNAs that emanate from alternative splicing of precursor mRNA in reversed order across exons. Despite the abundant presence of circRNAs in human genes and their involvement in diverse physiological processes, the functionality of most circRNAs remains a mystery. Like other non-coding RNAs, sub-cellular localization knowledge of circRNAs has the aptitude to demystify the influence of circRNAs on protein synthesis, degradation, destination, their association with different diseases, and potential for drug development. To date, wet experimental approaches are being used to detect sub-cellular locations of circular RNAs. These approaches help to elucidate the role of circRNAs as protein scaffolds, RNA-binding protein (RBP) sponges, micro-RNA (miRNA) sponges, parental gene expression modifiers, alternative splicing regulators, and transcription regulators. To complement wet-lab experiments, considering the progress made by machine learning approaches for the determination of sub-cellular localization of other non-coding RNAs, the paper in hand develops a computational framework, Circ-LocNet, to precisely detect circRNA sub-cellular localization. Circ-LocNet performs comprehensive extrinsic evaluation of 7 residue frequency-based, residue order and frequency-based, and physio-chemical property-based sequence descriptors using the five most widely used machine learning classifiers. Further, it explores the performance impact of K-order sequence descriptor fusion where it ensembles similar as well dissimilar genres of statistical representation learning approaches to reap the combined benefits. Considering the diversity of statistical representation learning schemes, it assesses the performance of second-order, third-order, and going all the way up to seventh-order sequence descriptor fusion. A comprehensive empirical evaluation of Circ-LocNet over a newly developed benchmark dataset using different settings reveals that standalone residue frequency-based sequence descriptors and tree-based classifiers are more suitable to predict sub-cellular localization of circular RNAs. Further, K-order heterogeneous sequence descriptors fusion in combination with tree-based classifiers most accurately predict sub-cellular localization of circular RNAs. We anticipate this study will act as a rich baseline and push the development of robust computational methodologies for the accurate sub-cellular localization determination of novel circRNAs.
Zusammenfassung Das starke Wachstum des E‑Commerce stellt bedeutende Herausforderungen an die Betreiber von Fahrzeugflotten, um den kostenintensiven Fahrzeugeinsatz in einem digital getriebenen Transportmarkt zu optimieren. Innovative Fahrzeug- und Datentechnologien bieten dabei neue Möglichkeiten für die serviceorientierte Gestaltung digitaler Ökosysteme mit weiteren Stakeholdern durch cloudbasierte Infrastrukturen. Bei genauerer Betrachtung der Transportkette wurde ein durchgehendes, automatisiertes und vernetztes Flottenmanagement in einem gemeinsamen Datenraum bisher nicht realisiert. In diesem Beitrag stellen die Autoren das Konzept „Smart Managed Freight Fleet“ vor, das im Rahmen des vom BMWK geförderten Konsortialprojekts „GAIA-X 4 ROMS“ (Remote Operation for Automated and Connected Mobility Services) entwickelt werden soll. Dazu zeigen die Autoren zunächst die bestehenden Informations- und Fahrzeugtechnologien für ein vernetztes Flottenmanagement auf. Anschließend wird ein innovativer multiagentenbasierter Flottenmanagementansatz beschrieben, der insbesondere Telematik-gestützte „intelligente Wechselbrücken- und Trailer (iWT)“ mit neuartigen „autonom navigierenden Paketstationen (ANP)“ auf der ersten bzw. letzten Meile verknüpft. Ein sicherer und souveräner Datenaustausch zwischen den Akteuren, Softwareagenten und weiteren Diensten wird dabei durch ein Gaia-X-konformes Datenökosystem erfolgen. Im Anschluss werden die hieraus entstehenden wissenschaftlichen und praktischen Implikationen für ein datenbasiertes Flottenmanagement beschrieben. Abschließend gibt der Artikel einen Ausblick auf die nächsten Entwicklungsschritte zur Gestaltung frei nutzbarer Flottenmanagementdienste in einem interoperablen Datenökosystem.
Viral-host protein protein interaction (PPI) analysis is essential to decode the molecular mechanism of viral pathogen and host immunity processes which eventually help to control viral diseases and optimize therapeutics. The state-of-the-art viral-host PPI predictor leverages unsupervised embedding learning technique (doc2vec) to generate statistical representations of viral-host protein sequences and a Random Forest classifier for interaction prediction. However, doc2vec approach generates the statistical representations of viral-host protein sequences by merely modelling the local context of residues which only partially captures residue semantics. The paper in hand proposes a novel technique for generating better statistical representations of viral and host protein sequences based on the infusion of comprehensive local and global contextual information of the residues. While local residue context aware encoding captures semantic relatedness and short range dependencies of residues. Global residue context aware encoding captures comprehensive long-range residues dependencies, positional invariance of residues, and unique residue combination distribution important for interaction prediction. Using concatenated rich statistical representations of viral and host protein sequences, a robust machine learning framework "LGCA-VHPPI" is developed which makes use of a deep forest model to effectively model complex non-linearity of viral-host PPI sequences. An in-depth performance comparison of the proposed LGCA-VHPPI framework with existing diverse sequence encoding schemes based viral-host PPI predictors reveals that LGCA-VHPPI outperforms state-of-the-art predictor by 6%, 2%, and 2% in terms of matthews correlation coefficient over 3 different benchmark viral-host PPI prediction datasets.
While sports activity recognition is a well studied subject in mobile, wearable and ubiquitous computing, work to date mostly focuses on recognition and counting of specific exercise types. Quality assessment is a much more difficult problem with significantly less published results. In this work, we present Quali-Mat: a method for evaluating the quality of execution (QoE) in exercises using a smart sports mat that can measure the dynamic pressure profiles during full-body, body-weight exercises. As an example, our system not only recognizes that the user is doing push-ups, but also distinguishes 5 subtly different types of push-ups, each of which (according to sports science literature and professional trainers) has a different effect on different muscle groups. We have investigated various machine learning algorithms targeting the specific type of spatio-temporal data produced by the pressure mat system. We demonstrate that computationally efficient, yet effective Conv3D model outperforms more complex state-of-the-art options such as transfer learning from the image domain. The approach is validated through an experiment designed to cover 47 quantifiable variants of 9 basic exercises with 12 participants. Overall, the model can categorize 9 exercises with 98.6% accuracy / 98.6% F1 score, and 47 QoE variants with 67.3% accuracy / 68.1% F1 score. Through extensive discussions with both the experiment results and practical sports considerations, our approach can be used for not only precisely recognizing the type of exercises, but also quantifying the workout quality of execution on a fine time granularity. We also make the Quali-Mat data set available to the community to encourage further research in the area.
Discrimination of circular RNA from long non-coding RNA is important to understand its role in different biological processes, disease prediction and cure. Identifying circular RNA through manual laboratories work is expensive, time-consuming and prone to errors. Development of computational methodologies for identification of circular RNA is an active area of research. State-of-the-art circular RNA identification methodologies make use of handcrafted features, which not only increase the feature space, but also extract irrelevant and redundant features. The paper in hand proposes an end-to-end deep learning-based framework named as CircNet, which does not require any handcrafted features. It takes raw RNA sequence as an input and utilises encoder–decoder based convolutional operations to learn lower-dimensional latent representation. This latent representation is further passed to another convolutional architecture to extract discriminative features followed by a classification layer. We performed extensive experimentation to highlight different regions of genome sequence that preserve the most important information for identifying circular RNAs. CircNet significantly outperforms state-of-the-art approaches with a considerable margin 10.29% in terms F1 measure.
The optimization of parallel kinematic manipulators (PKM) involve several constraints that are difficult to formalize, thus making optimal synthesis problem highly challenging. The presence of passive joint limits as well as the singularities and self-collisions lead to a complicated relation between the input and output parameters. In this article, a novel optimization methodology is proposed by combining a local search, Nelder–Mead algorithm, with global search methodologies such as low discrepancy distribution for faster and more efficient exploration of the optimization space. The effect of the dimension of the optimization problem and the different constraints are discussed to highlight the complexities of closed-loop kinematic chain optimization. The work also presents the approaches used to consider constraints for passive joint boundaries as well as singularities to avoid internal collisions in such mechanisms. The proposed algorithm can also optimize the length of the prismatic actuators and the constraints can be added in modular fashion, allowing to understand the impact of given criteria on the final result. The application of the presented approach is used to optimize two PKMs of different degrees of freedom.
Identifying objective and reliable markers to tailor diagnosis and treatment of psychiatric patients remains a challenge, as conditions like major depression, bipolar disorder, or schizophrenia are qualified by complex behavior observations or subjective self-reports instead of easily measurable somatic features. Recent progress in computer vision, speech processing and machine learning has enabled detailed and objective characterization of human behavior in social interactions. However, the application of these technologies to personalized psychiatry is limited due to the lack of sufficiently large corpora that combine multi-modal measurements with longitudinal assessments of patients covering more than a single disorder. To close this gap, we introduce Mephesto, a multi-centre, multi-disorder longitudinal corpus creation effort designed to develop and validate novel multi-modal markers for psychiatric conditions. Mephesto will consist of multi-modal audio-, video-, and physiological recordings as well as clinical assessments of psychiatric patients covering a six-week main study period as well as several follow-up recordings spread across twelve months. We outline the rationale and study protocol and introduce four cardinal use cases that will build the foundation of a new state of the art in personalized treatment strategies for psychiatric disorders.
Producing and consuming live-streamed content is a growing trend attracting many people today. While the actual content that is streamed is diverse, one especially popular context is games. Streamers of gaming content broadcast how they play digital or analog games, attracting several thousand viewers at once. Previous scientific work has revealed that different motivations drive people to become viewers, which apparently impacts how they interact with the offered features and which streamers’ behaviors they appreciate. In this paper, we wanted to understand whether viewers’ motivations can be formulated as viewer types and systematically measured. We present an exploratory factor analysis (followed by a validation study) with which we developed a 25-item questionnaire assessing five different viewer types. In addition, we analyzed the predictive validity of the viewer types for existing and potential live stream features. We were able to show that a relationship between the assessed viewer type and preferences for streamers’ behaviors and features in a stream exists, which can guide fellow researchers and streamers to understand viewers better and potentially provide more suitable experiences.
Cover Caption: The cover image is based on the Review Article Physical and mental well‐being of cobot workers: A scoping review using the Software‐Hardware‐Environment‐Liveware‐Liveware‐Organization model by Fabio A. Storm et al.,
Autonomous vehicle emergence with the potential to improve the traffic system efficiency and user comfort have made the co-existence of human-driven and autonomous vehicles inevitable in the near future. The different vehicle type co-existence has facilitated vehicle speed harmonisation to enhance traffic flow efficiency and prevent vehicle collision risk on the road. To a large extent, speed control and supervision of mix-traffic behaviours will go a long way to ameliorate the concerns envisaged in the autonomous vehicle integration process. A model predictive control-based autonomous vehicle speed adjustment technique with safe distance is developed to optimise the flow of mixed vehicles based on estimated driving behaviour. The main contribution of this work is employing the autonomous vehicle speed adjustment to the existing car-following model in mixed traffic. A mixed-traffic simulator is developed to test the proposed method in a car following model using a merging road to quantify the benefit of the proposed speed control strategy. The proposed simulation model is validated, and experiments are conducted with varying traffic intersection control strategies and vehicle type proportions. The obtained results demonstrate that the speed adjustment strategy has about 18.2% performance margin.
Medical image registration allows comparing images from different patients, modalities or time-points, but often suffers from missing correspondences due to pathologies and inter-patient variations.
Recent work suggests to explain trade-offs between soft-goals in terms of their conflicts, i.e., minimal unsolvable soft-goal subsets. But this does not explain the conflicts themselves: Why can a given set of soft-goals not be jointly achieved? Here we approach that question in terms of the underlying constraints on plans in the task at hand, namely resource availability and time windows. In this context, a natural form of explanation for a soft-goal conflict is a minimal constraint relaxation under which the conflict disappears (``if the deadline was 1 hour later, it would work''). We explore algorithms for computing such explanations. A baseline is to simply loop over all relaxed tasks and compute the conflicts for each separately. We improve over this by two algorithms that leverage information -- conflicts, reachable states -- across relaxed tasks. We show that these algorithms can exponentially outperform the baseline in theory, and we run experiments confirming that advantage in practice.
Citations are generally analyzed using only quantitative measures while excluding qualitative aspects such as sentiment and intent. However, qualitative aspects provide deeper insights into the impact of a scientific research artifact and make it possible to focus on relevant literature free from bias associated with quantitative aspects. Therefore, it is possible to rank and categorize papers based on their sentiment and intent. For this purpose, larger citation sentiment datasets are required. However, from a time and cost perspective, curating a large citation sentiment dataset is a challenging task. Particularly, citation sentiment analysis suffers from both data scarcity and tremendous costs for dataset annotation. To overcome the bottleneck of data scarcity in the citation analysis domain we explore the impact of out-domain data during training to enhance the model performance. Our results emphasize the use of different scheduling methods based on the use case. We empirically found that a model trained using sequential data scheduling is more suitable for domain-specific usecases. Conversely, shuffled data feeding achieves better performance on a cross-domain task. Based on our findings, we propose an end-to-end trainable multi-task model that covers the sentiment and intent analysis that utilizes out-domain datasets to overcome the data scarcity.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
537 members
Günter Neumann
  • Language Technology
Hans Uszkoreit
  • Language Technology
Serge Autexier
  • Cyber-Physical Systems
Philippe Thomas
  • Language Technology
Kaiserslautern, Germany