TY - JOUR A1 - Kamionka, Thomas A1 - Martens, Michael A1 - Chou, Kang Wei A1 - Curcic, Michael A1 - Drews, André A1 - Schütz, Gisela A1 - Tyliszczak, Tolek A1 - Stoll, Hermann A1 - Van Waeyenberge, Bartel A1 - Meier, Guido T1 - Magnetic Antivortex-Core Reversal by Circular-Rotational Spin Currents JF - Physical Review Letters Y1 - 2010 U6 - https://doi.org/10.1103/PhysRevLett.105.137204 SN - 0031-9007 VL - 105 IS - 13 PB - American Physical Society ER - TY - THES A1 - Stuhlsatz, André T1 - Machine learning with Lipschitz classifiers N2 - complex image, like for example that of a known person, and to distinguish it from other objects within half a second. While for a solution of this task the brain has access to a massive parallelism and a vast, hierarchically organized, and auto-associative memory, common computer architectures are just able to a sequential processing of information stored in a non auto-associative memory. Even modern, parallelly operating, multi-processor systems are far away from the performance of our brain. However, nowadays, it is possible to solve complex and memory extensive pattern recognition problems, like the recognition of handwritten digits or the transcription of speech, satisfactorily with a common computer by the use of modern statistical and algorithmic learning approaches. One of the most successful pattern recognition methods is the so-called Support Vector Machine (SVM). The SVM is based on the learning paradigm of structural risk minimization, which outperforms empirical approaches if only few data is available for solving the considered classification problem. Although the SVM has proven very good recognition performances in many cases, the SVM also comes up with limitations, for example if specific a priori knowledge shall be used. In particular, the increasing complexity of applications requires a high adaptivity of the classification method to the specific problem. Also concerning this point, the SVM is limited due to a restricted variety of implementable classification functions. The objective of the present thesis is the development of new learning algorithms for the classification of patterns, that on the one hand overcome the limitations of the SVM, but on the other hand are based on the same theoretical concepts facilitating the good performance of the SVM. Two new algorithms will be presented that are justified by a theoretical generalization of the SVM, and which will be utilized for the first time for a practical implementation. In contrast to the SVM, the new methods make accessible a much larger function class for constructing a classifier. This is an important prerequisite for flexible adaptation of the classifier to difficult classification tasks with particular requirements as well as for the integration of a priori knowledge about the problem at hand. In this work, the way to implementable algorithms leads across different mathematical reformulations of the original problem. Starting with the theoretical generalization of the SVM, it results a restricted optimization problem that is difficult to solve in general. In a first step, this problem is expressed in terms of a restricted minimax-problem by a modification of the suitable classification functions to a still very large function class consisting of (affine-)linear combinations of at least one-time continuously differentiable functions. In the next step, the minimax-problem is converted into a so-called Semi-Infinite Problem (SIP). It turns out, that this particular mathematical problem is appropriate in order to obtain a solution of the original problem for the considered function class using well-known optimization methods. To further exploit the problem structure, an equivalent dual problem is derived from the SIP. Therefore, we prove a duality theorem about the equality of the optimal values of the dual and the original problem. For solving the dual problem, a multilevel iterative approach is developed from which the proposed algorithms follow by pursuing different solution strategies. Moreover, all sub-optimization methods of any stage necessary for an implementation in software are developed. Namely, these are an adapted interior-point-method, a simulated annealing based search heuristics and a particular gradient decent approach. Furthermore, options are depicted for an improvement of efficiency for future implementations. Besides the emphasis on the theoretical development of new learning methods and their practical implementations, all algorithms were implemented in the MATLAB(R) programming environment for the experimental part of the present thesis. Hence, they are also available for further research purposes in future. For the first time, classification results are explored and evaluated in comparison to the SVM on different data sets. As test data, an artificial 2d-dataset as well as two real-world datasets were used. In the concluding experiment, a scenario is prototypically considered to which the SVM is only inadequately applicable and which shall precisely prove the capability of the new methods in that case. It follows, regarding the considered datasets, the proposed learning methods reach comparably good classification accuracy like the SVM in standard applications. Moreover, the particular benefit of the new methods is reflected theoretically and experimentally in the ability to solve classification problems using decision functions that are not accessible to SVMs. Thereby, the underlying ideas, which make the SVM excel compared to other approaches with respect to generalization performances in case of few available learning information, are adequately transported into the proposed new environment. This opens the way for a design and a use of new classifiers that have not been implementable in a robust and generalizing basic concept so far. Y1 - 2010 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:101:1-201104192945 PB - Otto-von-Guericke-Universität Magdeburg CY - Magdeburg ER - TY - CHAP A1 - Stöhr, Marcel A1 - Zielke, Thomas ED - Leipzig, Konrad von ED - Sacks, Natasha ED - Mc Clelland, Michelle T1 - Machine Learning for Soft Sensors and an Application in Cement Production T2 - Smart, Sustainable Manufacturing in an Ever-Changing World Proceedings of International Conference on Competitive Manufacturing (COMA ’22) KW - Soft sensors KW - Machine learning KW - Dynamic processes KW - Cement production KW - Robustness KW - Artificial intelligence Y1 - 2023 SN - 978-3-031-15604-5 U6 - https://doi.org/10.1007/978-3-031-15602-1_46 VL - Lecture Notes in Production Engineering SP - 627 EP - 638 PB - Springer Nature Switzerland CY - Cham ER - TY - THES A1 - Krieter, Philipp T1 - Looking Inside - Mobile Screen Recordings as a Privacy Friendly Long-Term Data Source to Analyze User Behavior N2 - Mobile devices are ubiquitous in many societies and shape the way we interact with technology and each other. Research on how we use and perceive technology is essential to understand its impact. This work advances how we can follow user behavior on mobile devices. We combine the strength of two common data sources for tracking on mobile devices, log files, and screen recordings. Log files are suitable for long-term and privacy-friendly analyzation but provide rather general data (e.g. system log files) unless one has access to the source code of the applications or operating systems. Screen recordings are usually used for short-termed analysis (e.g. usability tests) because the analysis is time-consuming, but they provide all activities on the screen in high detail regardless of which application or operating system. This thesis combines both data sources and presents an approach to automatically generate log files from mobile screen recordings. The approach utilizes methods of computer vision and machine learning to automatically process screen recordings and extend their use. Screen recordings reveal virtually everything a user does with a device, making privacy important, especially in user studies. We present a privacy concept and implementation and show how the risk of exposing private data can be reduced, by processing all recordings locally on the mobile devices and anonymizing the resulting log files. In order to apply the developed method in practice, we carry out a study in the context of education and show how log files of screen recordings can complement and extend existing research in learning analytics. This thesis opens up novel perspectives on how we can look at human-computer interaction with mobile devices. We show how to generate long-term log data with high detail and accuracy from mobile screen recordings, in a privacy-friendly way, locally on mobile devices. KW - mobile screen recordings KW - log files KW - learning analytics KW - user behavior KW - privacy Y1 - 2020 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:gbv:46-elib43189 PB - Bremen CY - Universität Bremen ER - TY - CHAP A1 - Steffens, Jochen A1 - Guastavino, Catherine T1 - Looking back by looking into the future – The role of anticipation and trend effects in retrospective judgments of musical excitement T2 - Proceedings of the Congress of the International Association of Empirical Aesthetics, New York City (NY), USA KW - Habilitation KW - Peak & End Y1 - 2014 CY - New York ER - TY - JOUR A1 - Vocks, Silja A1 - Wächter, Alexandra A1 - Wucherer, Maike A1 - Kosfelder, Joachim T1 - Look at yourself: can body image therapy affect the cognitive and emotional response to seeing oneself in the mirror in eating disorders? JF - European eating disorders review : the journal of the Eating Disorders Association N2 - The study examined whether the extent of negative body-related thoughts and emotions evoked by looking in a mirror can be changed by cognitive-behavioural body image therapy (BIT) and whether any remaining cognitive and affective responses dropped faster after the treatment. Females with eating disorders (n = 17) looked in a mirror for 40 minutes under standardised conditions before and after BIT. A control group (CG) (n = 24) was also exposed to their own bodies twice. Cognitive and affective reactions were assessed four times during the session. Hierarchical Linear Modelling revealed that among eating-disordered patients, the entrance levels of negative thoughts and feelings were lower at post compared to pre-BIT, whereas they remained stable in the CG. The extent of reduction of the remaining body image distress was not changed after the treatment. BIT seems to be a potent treatment of negative emotions and cognitions in an in vivo situation. KW - FBSK_Webseite Y1 - 2008 UR - http://www.ncbi.nlm.nih.gov/pubmed/17721910 U6 - https://doi.org/10.1002/erv.825 VL - 16 IS - 2 SP - 147 EP - 154 PB - Wiley ER - TY - CHAP A1 - Bögelein, Silke A1 - Brinkmann, Fabian A1 - Ackermann, David A1 - Weinzierl, Stefan T1 - Localization Cues of a Spherical Head Model [Abstract] T2 - Fortschritte der Akustik – DAGA 2018 - 44. Jahrestagung für Akustik 19.-22. März 2018, München Y1 - 2018 UR - https://pub.dega-akustik.de/DAGA_2018/data/articles/000459.pdf SN - 978-3-939296-13-3 SP - 347 EP - 350 PB - Deutsche Gesellschaft für Akustik e.V. ER - TY - JOUR A1 - Müller, Patrick A1 - Braun, Alexander T1 - Local performance evaluation of AI-algorithms with the generalized spatial recall index JF - tm - Technisches Messen Y1 - 2023 U6 - https://doi.org/10.1515/teme-2023-0013 SN - 2196-7113 VL - 90 IS - 7-8 SP - 464 EP - 477 PB - De Gruyter ER - TY - CHAP A1 - Macklin, Graham A1 - Virchow, Fabian ED - Macklin, Graham ED - Copsey, Nigel T1 - Local embedding as a factor in electoral success: the BNP and NPD compared T2 - The British National Party KW - Forena Y1 - 2011 SN - 9781136820625 SP - 1 EP - 20 PB - Routledge CY - Abingdon, Oxon; New York ER - TY - CHAP A1 - Blättermann, Patrick A1 - Versümer, Siegbert A1 - Steffens, Jochen T1 - Loadings of Acoustical Metrics on Soundscape Items and their relation to Soundscape Dimensions across different datasets [Abstract] T2 - DAS|DAGA 2025: 51st Annual Meeting on Acoustics, March 17-20, 2025, Copenhagen KW - Soundscape KW - Psychoakustik KW - Datenanalyse KW - Maschinelles Lernen Y1 - 2025 UR - https://pub.dega-akustik.de/DAS-DAGA_2025/konferenz-2007.html?article=383 PB - Deutsche Gesellschaft für Akustik e.V. CY - Berlin ER -