TY - INPR A1 - de Jonge, Niek F. A1 - Joas, David A1 - Truong, Lem-Joe A1 - van der Hooft, Justin J.J. A1 - Huber, Florian T1 - Reliable cross-ion mode chemical similarity prediction between MS2 spectra T2 - biorxiv N2 - Mass spectrometry is commonly used to characterize metabolites in untargeted metabolomics. This can be done in positive and negative ionization mode, a choice typically guided by the fraction of metabolites a researcher is interested in. During analysis, mass spectral comparisons are widely used to enable annotation through reference libraries and to facilitate data organization through networking. However, until now, such comparisons between mass spectra were restricted to mass spectra of the same ionization mode, as the two modes generally result in very distinct fragmentation spectra. To overcome this barrier, here, we have implemented a machine learning model that can predict chemical similarity between spectra of different ionization modes. Hence, our new MS2DeepScore 2.0 model facilitates the seamless integration of positive and negative ionization mode mass spectra into one analysis pipeline. This creates entirely new options for data exploration, such as mass spectral library searching of negative ion mode spectra in positive ion mode libraries or cross-ionization mode molecular networking. Furthermore, to improve the reliability of predictions and better cope with unseen data, we have implemented a method to estimate the quality of prediction. This will help to avoid false predictions on spectra with low information content or spectra that substantially differ from the training data. We anticipate that the MS2DeepScore 2.0 model will extend our current capabilities in organizing and annotating untargeted metabolomics profiles. KW - Massenspektrometrie KW - Metabolomik KW - Maschinelles Lernen Y1 - 2024 U6 - https://doi.org/10.1101/2024.03.25.586580 PB - Cold Spring Harbor Laboratory ER - TY - INPR A1 - Yuwono, Steve A1 - Schwung, Dorothea A1 - Schwung, Andreas T1 - Transfer learning of state-based potential games for process optimization in decentralized manufacturing systems T2 - arXiv N2 - This paper presents a novel transfer learning approach in state-based potential games (TL-SbPGs) for enhancing distributed self-optimization in manufacturing systems. The approach focuses on the practical relevant industrial setting where sharing and transferring gained knowledge among similar-behaved players improves the self-learning mechanism in large-scale systems. With TL-SbPGs, the gained knowledge can be reused by other players to optimize their policies, thereby improving the learning outcomes of the players and accelerating the learning process. To accomplish this goal, we develop transfer learning concepts and similarity criteria for players, which offer two distinct settings: (a) predefined similarities between players and (b) dynamically inferred similarities between players during training. We formally prove the applicability of the SbPG framework in transfer learning. Additionally, we introduce an efficient method to determine the optimal timing and weighting of the transfer learning procedure during the training phase. Through experiments on a laboratory-scale testbed, we demonstrate that TL-SbPGs significantly boost production efficiency while reducing power consumption of the production schedules while also outperforming native SbPGs. KW - Maschinelles Lernen KW - Spieltheorie KW - Produktionssystem KW - Optimierung Y1 - 2024 U6 - https://doi.org/10.48550/arXiv.2408.05992 PB - arXiv ER - TY - INPR A1 - Yuwono, Steve A1 - Löppenberg, Marlon A1 - Schwung, Dorothea A1 - Schwung, Andreas T1 - Gradient-based Learning in State-based Potential Games for Self-Learning Production Systems T2 - arxiv N2 - In this paper, we introduce novel gradient-based optimization methods for state-based potential games (SbPGs) within self-learning distributed production systems. SbPGs are recognised for their efficacy in enabling self-optimizing distributed multi-agent systems and offer a proven convergence guarantee, which facilitates collaborative player efforts towards global objectives. Our study strives to replace conventional ad-hoc random exploration-based learning in SbPGs with contemporary gradient-based approaches, which aim for faster convergence and smoother exploration dynamics, thereby shortening training duration while upholding the efficacy of SbPGs. Moreover, we propose three distinct variants for estimating the objective function of gradient-based learning, each developed to suit the unique characteristics of the systems under consideration. To validate our methodology, we apply it to a laboratory testbed, namely Bulk Good Laboratory Plant, which represents a smart and flexible distributed multi-agent production system. The incorporation of gradient-based learning in SbPGs reduces training times and achieves more optimal policies than its baseline. KW - Machine Learning KW - Artificial Intelligence KW - Computer Science and Game Theory KW - Maschinelles Lernen KW - Künstliche Intelligenz KW - Spieltheorie Y1 - 2024 U6 - https://doi.org/10.48550/arXiv.2406.10015 PB - arxiv ER - TY - INPR A1 - Bushuiev, Roman A1 - Bushuiev, Anton A1 - de Jonge, Niek F. A1 - Young, Adamo A1 - Kretschmer, Fleming A1 - Samusevich, Raman A1 - Heirman, Janne A1 - Wang, Fei A1 - Zhang, Luke A1 - Dührkop, Kai A1 - Ludwig, Marcus A1 - Haupt, Nils A. A1 - Kalia, Apurva A1 - Brungs, Corinna A1 - Schmid, Robin A1 - Greiner, Russell A1 - Wang, Bo A1 - Wishart, David S. A1 - Liu, Li-Ping A1 - Rousu, Juho A1 - Bittremieux, Wout A1 - Röst, Hannes A1 - Mak, Tytus D. A1 - Hassoun, Soha A1 - Huber, Florian A1 - van der Hooft, Justin J.J. A1 - Stravs, Michael A. A1 - Böcker, Sebastian A1 - Sivic, Josef A1 - Pluskal, Tomáš T1 - MassSpecGym: A benchmark for the discovery and identification of molecules T2 - arXiv N2 - The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a result, the vast majority of acquired MS/MS spectra remain uninterpreted, thereby limiting our understanding of the underlying (bio)chemical processes. Despite decades of progress in machine learning applications for predicting molecular structures from MS/MS spectra, the development of new methods is severely hindered by the lack of standard datasets and evaluation protocols. To address this problem, we propose MassSpecGym -- the first comprehensive benchmark for the discovery and identification of molecules from MS/MS data. Our benchmark comprises the largest publicly available collection of high-quality labeled MS/MS spectra and defines three MS/MS annotation challenges: de novo molecular structure generation, molecule retrieval, and spectrum simulation. It includes new evaluation metrics and a generalization-demanding data split, therefore standardizing the MS/MS annotation tasks and rendering the problem accessible to the broad machine learning community. MassSpecGym is publicly available at this https URL [https://github.com/pluskal-lab/MassSpecGym]. KW - Maschinelles Lernen KW - Benchmark KW - Computational chemistry KW - Massenspektrometrie KW - Molekülstruktur Y1 - 2025 U6 - https://doi.org/10.48550/arXiv.2410.23326 N1 - Open Review: https://web.archive.org/web/20250521010648/https://openreview.net/forum?id=AAo8zAShX3#discussion PB - arXiv ET - v3 ER - TY - INPR A1 - Müller, Patrick A1 - Braun, Alexander A1 - Keuper, Margret T1 - Examining the Impact of Optical Aberrations to Image Classification and Object Detection Models N2 - Deep neural networks (DNNs) have proven to be successful in various computer vision applications such that models even infer in safety-critical situations. Therefore, vision models have to behave in a robust way to disturbances such as noise or blur. While seminal benchmarks exist to evaluate model robustness to diverse corruptions, blur is often approximated in an overly simplistic way to model defocus, while ignoring the different blur kernel shapes that result from optical systems. To study model robustness against realistic optical blur effects, this paper proposes two datasets of blur corruptions, which we denote OpticsBench and LensCorruptions. OpticsBench examines primary aberrations such as coma, defocus, and astigmatism, i.e. aberrations that can be represented by varying a single parameter of Zernike polynomials. To go beyond the principled but synthetic setting of primary aberrations, LensCorruptions samples linear combinations in the vector space spanned by Zernike polynomials, corresponding to 100 real lenses. Evaluations for image classification and object detection on ImageNet and MSCOCO show that for a variety of different pre-trained models, the performance on OpticsBench and LensCorruptions varies significantly, indicating the need to consider realistic image corruptions to evaluate a model's robustness against blur. KW - Maschinelles Lernen KW - Maschinelles Sehen KW - Tiefes neuronales Netz KW - Objekterkennung KW - Abbildungsfehler Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:hbz:due62-opus-55457 PB - arXiv ET - v1 ER - TY - INPR A1 - Yuwono, Steve A1 - Schwung, Dorothea A1 - Schwung, Andreas T1 - Distributed Stackelberg Strategies in State-based Potential Games for Autonomous Decentralized Learning Manufacturing Systems N2 - This article describes a novel game structure for autonomously optimizing decentralized manufacturing systems with multi-objective optimization challenges, namely Distributed Stackelberg Strategies in State-Based Potential Games (DS2-SbPG). DS2-SbPG integrates potential games and Stackelberg games, which improves the cooperative trade-off capabilities of potential games and the multi-objective optimization handling by Stackelberg games. Notably, all training procedures remain conducted in a fully distributed manner. DS2-SbPG offers a promising solution to finding optimal trade-offs between objectives by eliminating the complexities of setting up combined objective optimization functions for individual players in self- learning domains, particularly in real-world industrial settings with diverse and numerous objectives between the sub-systems. We further prove that DS2-SbPG constitutes a dynamic potential game that results in corresponding converge guarantees. Experimental validation conducted on a laboratory-scale testbed highlights the efficacy of DS2-SbPG and its two variants, such as DS2-SbPG for single-leader-follower and Stack DS2-SbPG for multi-leader-follower. The results show significant reductions in power consumption and improvements in overall performance, which signals the potential of DS2-SbPG in real-world applications. KW - Maschinelles Lernen KW - Produktionssystem KW - Spieltheorie KW - Optimierung Y1 - 2024 U6 - https://doi.org/10.48550/arXiv.2408.06397 PB - arXiv ER -