PUMA publications for /tag/MachineLearningTue Feb 17 15:29:11 CET 2026Capturing reduced-order quantum many-body dynamics out of equilibrium via neural ordinary differential equations2025IRIS machinelearning manybodyml myown patrickegenlauf physics publist Thu Jan 08 10:27:23 CET 2026Capturing reduced-order quantum many-body dynamics out of equilibrium via neural ordinary differential equations2025myown patrickegenlauf publist manybodyml physics machinelearning Thu Dec 11 10:23:38 CET 2025OctoberEnhancing Spaceflight Autonomy with Machine Learning: Fault Detection in Life Support Systems2025MachineLearning V-HAB hsf irs myown Wed Dec 10 11:18:04 CET 2025Optics & Laser Technologydec113441Multimodal sensor fusion with SWIR imaging and audio for inline gas-tightness monitoring in laser-welded bipolar plates1922025myown peer machinelearning laserwelding sensorfusion Wed Dec 10 11:14:41 CET 2025Optics & Laser Technologydec113441Multimodal sensor fusion with SWIR imaging and audio for inline gas-tightness monitoring in laser-welded bipolar plates1922025laserwelding machinelearning myown peer sensorfusion Thu Sep 11 14:56:36 CEST 2025The EPS ForummarCoarse-graining non-equilibrium systems with machine learning: from conceptual challenges to new approaches2024machinelearning manybodyml myown patrickegenlauf physics posterlist Wed Sep 10 16:04:33 CEST 2025DPG Spring Meetings 2024marCoarse-graining non-equilibrium systems with machine learning2024machinelearning manybodyml myown patrickegenlauf physics talklist Wed May 07 16:32:35 CEST 2025International Journal of Production Researchaug51692–1706A methodology for evaluating feature selection and clustering methods with project-specific requirements632024MachineLearning isw This paper describes amethodology for ranking feature selection and clusteringmethods with userspecific preferences and taking data properties into account. For a better understanding of this paper, the developed methodology is referred to as the Two Machine Learning Procedures, Preferences and Properties (2ML3P) methodology. The 2ML3P methodology aims to support users from multiple domains, such as engineers, who have little expertise in machine learning (ML). It is also independent from the disciplinary core competencies of the manufacturer, with a strong focus on employability in small andmid-sized enterprises (SME). The foundationof themethodology toevaluate the combination of the twomachine learning classes is described. It focuses on a range of feature selection and clustering methods, their limitations, and their challenges. The paper covers the concept phase by defining the inputs, such as the specific characteristics ofmachine learning classes or the properties of the production data and the user preferences. With applied methodologies such as the analytic hierarchy process (AHP) and the technique for order preference by similarity to ideal solution (TOPSIS), the preferences of the user as valid input are integrated. The scientific contribution of this methodology is the approach to include user preferences and specific data properties in the selection process of twoMLmethods. As digitalisation progresses, making data-driven decisions in the domains of production and logistics is a goal for many SMEs. This methodology can support a data-driven decision-aid model by providing a guided method, which requires relatively little ML knowledge on the part of the engineer. It allows the user(s) to select the best suited combination of ML methods for a clustering use case.Mon Oct 21 15:34:45 CEST 2024Machine Learning: Science and Technology03Gaussian-process-regression-based method for the localization of exceptional points in complex resonance spectra2024MachineLearning PatrickEgenlauf QuantumMechanics myown patrickegenlauf physics publist 01Thu May 23 11:30:52 CEST 2024J. Manuf. Mater. Process.107A Data-Driven Approach for Cutting Force Prediction in FEM Machining Simulations Using Gradient Boosted Machines8(3)2024FEM MachineLearning machining manufacturing Thu May 23 11:30:52 CEST 2024J. Manuf. Mater. Process.107A Data-Driven Approach for Cutting Force Prediction in FEM Machining Simulations Using Gradient Boosted Machines8(3)2024machining manufacturing FEM MachineLearning Wed Jan 18 08:53:14 CET 2023janPreprint!Ontology Extension with {NLP}-based Concept Extraction for Domain Experts in Catalytic Sciences2023forschungsdaten machinelearning metadata ontologie Mon Oct 10 09:56:21 CEST 2022DatasetRelated to: Baier, A., Boukhers, Z., & Staab, S. (2021). Hybrid Physics and Deep Learning Model for Interpretable Vehicle State Prediction. arXiv: 2103.06727A Simulated 4-DOF Ship Motion Dataset for System Identification under Environmental Disturbances2022EXC2075 deeplearning exc2075 machinelearning myown pn4 simtech This dataset contains data of 125 1-hour simulations of ship motion during various sea states performing random maneuvers in 4 degrees of freedom (surge-sway-yaw-roll). The original ship is a patrol ship developed by Perez et al. [1]. We have extended it with a set of two symmetrically placed rudder propellers. Additionally, we simulate wind forces according to Isherwood's wind model [2]. Wind-induced waves are generated with the JONSWAP spectrum [3] and the corresponding wave forces are then computed using wave force response amplitude operators (ROA).Implementations of the ship model, Isherwood's wave model, wave force ROAs and the JONSWAP spectrum can be found in the Marine Systems Simulator toolbox by Fossen and Perez [4].The dataset is split into a routine operation set (96 hours) and into an Out-Of-Distribution (OOD) set (29 hours). The routine operation set is split into train-validation-test with a 60-10-30 split, while the OOD set is used solely for testing.The dataset is used for the evaluation of nonlinear system identification methods for multi-step predictions. The following inputs and outputs are considered for the identification problem. Inputs are the shaft speeds of both propellers, their azimut angles, wind angle of attack, and wind speed. Measured states or outputs are surge velocity, sway velocity and roll rate, as well as yaw angle and roll angle.Please see the README.txt file for details regarding the file structure of this dataset and a description of the variables in the .tab files.This research is funded by Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy - EXC 2075 - 390740016. We acknowledge the support by the Stuttgart Center for Simulation Science (SimTech).[1] T. Perez, A. Ross, and T. I. Fossen, “A 4-DOF SIMULINK model of acoastal patrol vessel for manoeuvring in waves,” in IFAC MCMC, 2006.[2] R. M. Isherwood, “Wind resistance of merchant ships,” The RoyalInstitution of Naval Architects, 1972.[3] K. Hasselmann, T. Barnett, E. Bouws, H. Carlson, D. Cartwright, K. Enke,J. Ewing, H. Gienapp, D. Hasselmann, P. Kruseman, A. Meerburg,P. Muller, D. Olbers, K. Richter, W. Sell, and H. Walden, “Measurementsof wind-wave growth and swell decay during the joint north sea waveproject (JONSWAP),” Deut. Hydrogr. Z., vol. 8, pp. 1-95, 01 1973.[4] T. I. Fossen and T. Perez, “Marine Systems Simulator (MSS),” https://github.com/cybergalactic/MSS, 2004, last accessed: 2022-06-14.Mon Oct 10 09:56:21 CEST 2022DatasetRelated to: Baier, A., Boukhers, Z., & Staab, S. (2021). Hybrid Physics and Deep Learning Model for Interpretable Vehicle State Prediction. arXiv: 2103.06727A Simulated 4-DOF Ship Motion Dataset for System Identification under Environmental Disturbances2022myown simtech deeplearning pn4 from:alexbaier machinelearning EXC2075 exc2075 This dataset contains data of 125 1-hour simulations of ship motion during various sea states performing random maneuvers in 4 degrees of freedom (surge-sway-yaw-roll). The original ship is a patrol ship developed by Perez et al. [1]. We have extended it with a set of two symmetrically placed rudder propellers. Additionally, we simulate wind forces according to Isherwood's wind model [2]. Wind-induced waves are generated with the JONSWAP spectrum [3] and the corresponding wave forces are then computed using wave force response amplitude operators (ROA).Implementations of the ship model, Isherwood's wave model, wave force ROAs and the JONSWAP spectrum can be found in the Marine Systems Simulator toolbox by Fossen and Perez [4].The dataset is split into a routine operation set (96 hours) and into an Out-Of-Distribution (OOD) set (29 hours). The routine operation set is split into train-validation-test with a 60-10-30 split, while the OOD set is used solely for testing.The dataset is used for the evaluation of nonlinear system identification methods for multi-step predictions. The following inputs and outputs are considered for the identification problem. Inputs are the shaft speeds of both propellers, their azimut angles, wind angle of attack, and wind speed. Measured states or outputs are surge velocity, sway velocity and roll rate, as well as yaw angle and roll angle.Please see the README.txt file for details regarding the file structure of this dataset and a description of the variables in the .tab files.This research is funded by Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy - EXC 2075 - 390740016. We acknowledge the support by the Stuttgart Center for Simulation Science (SimTech).[1] T. Perez, A. Ross, and T. I. Fossen, “A 4-DOF SIMULINK model of acoastal patrol vessel for manoeuvring in waves,” in IFAC MCMC, 2006.[2] R. M. Isherwood, “Wind resistance of merchant ships,” The RoyalInstitution of Naval Architects, 1972.[3] K. Hasselmann, T. Barnett, E. Bouws, H. Carlson, D. Cartwright, K. Enke,J. Ewing, H. Gienapp, D. Hasselmann, P. Kruseman, A. Meerburg,P. Muller, D. Olbers, K. Richter, W. Sell, and H. Walden, “Measurementsof wind-wave growth and swell decay during the joint north sea waveproject (JONSWAP),” Deut. Hydrogr. Z., vol. 8, pp. 1-95, 01 1973.[4] T. I. Fossen and T. Perez, “Marine Systems Simulator (MSS),” https://github.com/cybergalactic/MSS, 2004, last accessed: 2022-06-14.Thu May 19 16:58:09 CEST 2022KZfSS K{\"o}lner Zeitschrift f{\"u}r Soziologie und SozialpsychologiemayApplying Machine Learning in Sociology: How to Predict Gender and Reveal Research Preferences2022machinelearning s7 sociology sowi_VII 19Applications of machine learning (ML) in industry and natural sciences yielded some of the most impactful innovations of the last decade (for instance, artificial intelligence, gene prediction or search engines) and changed the everyday-life of many people. From a methodological perspective, we can differentiate between unsupervised machine learning (UML) and supervised machine learning (SML). While SML uses labeled data as input to train algorithms in order to predict outcomes of unlabeled data, UML detects underlying patterns in unlabeled observations by exploiting the statistical properties of the data. The possibilities of ML for analyzing large datasets are slowly finding their way into the social sciences; yet, it lacks systematic introductions into the epistemologically alien subject. I present applications of some of the most common methods for SML (i.e., logistic regression) and UML (i.e., topic models). A practical example offers social scientists a ``how-to'' description for utilizing both. With regard to SML, the case is made by predicting gender of a large dataset of sociologists. The proposed approach is based on open-source data and outperforms a popular commercial application (genderize.io). Utilizing the predicted gender in topic models reveals the stark thematic differences between male and female scholars that have been widely overlooked in the literature. By applying ML, hence, the empirical results shed new light on the longstanding question of gender-specific biases in academia.Fri Nov 12 09:05:06 CET 2021cite arxiv:2103.06727Hybrid Physics and Deep Learning Model for Interpretable Vehicle State Prediction2021EXC2075 deeplearning machinelearning myown pn4 prePrint simtech Physical motion models offer interpretable predictions for the motion of vehicles. However, some model parameters, such as those related to aero- and hydrodynamics, are expensive to measure and are often only roughly approximated reducing prediction accuracy. Recurrent neural networks achieve high prediction accuracy at low cost, as they can use cheap measurements collected during routine operation of the vehicle, but their results are hard to interpret. To precisely predict vehicle states without expensive measurements of physical parameters, we propose a hybrid approach combining deep learning and physical motion models including a novel two-phase training procedure. We achieve interpretability by restricting the output range of the deep neural network as part of the hybrid model, which limits the uncertainty introduced by the neural network to a known quantity. We have evaluated our approach for the use case of ship and quadcopter motion. The results show that our hybrid model can improve model interpretability with no decrease in accuracy compared to existing deep learning approaches.Thu Oct 08 17:45:01 CEST 2020Proceedings of the 2020 Conference on Empirical Methods in Natural Language ProcessingDissecting Span Identification Tasks with Performance Prediction2020machinelearning metalearning myown ner nlp Tue Jun 09 11:14:14 CEST 20202019 Second International Conference on Artificial Intelligence for Industries (AI4I)Sep.79-84Reinforcement Learning of a Robot Cell Control Logic using a Software-in-the-Loop Simulation as Environment2019MachineLearning grk2198 isw myown This paper introduces a method for automatic robot programming of industrial robots using reinforcement learning on a Software-in-the-loop simulation. The focus of the the paper is on the higher levels of a hierarchical robot programming problem. While the lower levels the skills are stored as domain specific program code, the combination of the skills into a robot control program to solve a specific task is automated. The reinforcement learning learning approach allows the shopfloor workers and technicians just to define the end result of the manufacturing process through a reward function. The programming and process optimization is done within the learning procedure. The Software-in-the-loop simulation with the robot control software makes it possible to to interpret the real program code and generate the exact motion. The exact motion of the robot is needed in order to find not just an optimal but also a collision-free policy.Wed May 15 01:21:27 CEST 20192018 First International Conference on Artificial Intelligence for Industries (AI4I)Sep.77-80Reinforcement Learning of Material Flow Control Logic Using Hardware-in-the-Loop Simulation2018DigitalTwin MachineLearning grk2198 isw myown In this paper the concept of reinforcement learning agent is presented, which can deduce the correct control policy of a plant by acting in its digital twin (the HiL simulation). This way the agent substitutes a real control system. By using reinforcement learning methods, a proof of concept application is presented for a simplistic material flow system, with the same type of access to the digital twin which a PLC controller-hardware would have. With the presented approach the agent is able to find the correct control policy.Thu Jan 17 22:17:56 CET 20192018 25th International Conference on Mechatronics and Machine Vision in Practice (M2VIP)nov1-6Digital Twins of Manufacturing Systems as a Base for Machine Learning2018DigitalTwin DigitaleFabrik MachineLearning grk2198 isw myown In the engineering phase of modern manufacturing systems, simulation-based methods and tools have been established to face the increasing demands on time-efficiency and profitability. In the application of these simulation solutions, model-based digital twins are created, as multi-domain simulation models to describe the behavior of the manufacturing system. During the production process, a data-driven digital twin arises in the context of industry 4.0 based on an increasing networking and new cloud technologies. Recent developments in machine learning of fer new possibilities in conjunction with the digital twin. These range from data-based learning of models to learning control logic of complex systems. This paper proposes a combined model-based and data-driven concept of a digital twin. It shows how to use machine learning in connection with these models, in order to archive faster development times of manufacturing systems.MachineLearningCommunity for tag(s) MachineLearning