@inproceedings{StuhlsatzLippelZielke2010, author = {Stuhlsatz, Andr{\´e} and Lippel, Jens and Zielke, Thomas}, title = {Feature Extraction for Simple Classification}, series = {2010 20th International Conference on Pattern Recognition, August 23 - 26, 2010 , Istanbul, Turkey}, booktitle = {2010 20th International Conference on Pattern Recognition, August 23 - 26, 2010 , Istanbul, Turkey}, publisher = {IEEE}, isbn = {978-0-7695-4109-9}, issn = {1051-4651}, doi = {10.1109/ICPR.2010.377}, pages = {1525 -- 1528}, year = {2010}, language = {en} } @inproceedings{StuhlsatzMeyerEybenetal.2011, author = {Stuhlsatz, Andr{\´e} and Meyer, Christine and Eyben, Florian and Zielke, Thomas and Meier, Gunter and Schuller, Bjorn}, title = {Deep neural networks for acoustic emotion recognition: Raising the benchmarks}, series = {2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, booktitle = {2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, publisher = {IEEE}, doi = {10.1109/ICASSP.2011.5947651}, pages = {5688 -- 5691}, year = {2011}, abstract = {Deep Neural Networks (DNNs) denote multilayer artificial neural networks with more than one hidden layer and millions of free parameters. We propose a Generalized Discriminant Analysis (GerDA) based on DNNs to learn discriminative features of low dimension optimized with respect to a fast classification from a large set of acoustic features for emotion recognition. On nine frequently used emotional speech corpora, we compare the performance of GerDA features and their subsequent linear classification with previously reported benchmarks obtained using the same set of acoustic features classified by Support Vector Machines (SVMs). Our results impressively show that low-dimensional GerDA features capture hidden information from the acoustic features leading to a significantly raised unweighted average recall and considerably raised weighted average recall.}, language = {en} } @inproceedings{StoehrZielke2023, author = {St{\"o}hr, Marcel and Zielke, Thomas}, title = {Machine Learning for Soft Sensors and an Application in Cement Production}, series = {Smart, Sustainable Manufacturing in an Ever-Changing World Proceedings of International Conference on Competitive Manufacturing (COMA '22)}, volume = {Lecture Notes in Production Engineering}, booktitle = {Smart, Sustainable Manufacturing in an Ever-Changing World Proceedings of International Conference on Competitive Manufacturing (COMA '22)}, editor = {Leipzig, Konrad von and Sacks, Natasha and Mc Clelland, Michelle}, publisher = {Springer Nature Switzerland}, address = {Cham}, isbn = {978-3-031-15604-5}, doi = {10.1007/978-3-031-15602-1_46}, pages = {627 -- 638}, year = {2023}, language = {en} } @inproceedings{ZielkeBernstCrombachetal.2018, author = {Zielke, Thomas and Bernst, Irina and Crombach, Tobias and Becker, Martin and Lippel, Jens}, title = {WHAT CAN BE LEARNED BY GRADING ANALYTICS?}, series = {INTED 2018: 12th International Technology, Education and Development Conference : 5-7 March, 2018, Valencia (Spain) : conference proceedings}, volume = {12th International Technology, Education and Development Conference}, booktitle = {INTED 2018: 12th International Technology, Education and Development Conference : 5-7 March, 2018, Valencia (Spain) : conference proceedings}, editor = {G{\´o}mez Chova, Luis and L{\´o}pez Mart{\´i}nez, A. and Candel Torres, I.}, publisher = {IATED Academy}, address = {Valencia, Spain}, isbn = {978-84-697-9480-7}, doi = {10.21125/inted.2018.0593}, pages = {3097 -- 3106}, year = {2018}, language = {en} } @inproceedings{LippelBeckerZielke2019, author = {Lippel, Jens and Becker, Martin and Zielke, Thomas}, title = {Modeling Dynamic Processes with Deep Neural Networks: A Case Study with a Gas-fired Absorption Heat Pump}, series = {SIMULTECH 2019: Proceedings of the 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications : Prague, Czech RepublicJuly 29-31, 2019}, booktitle = {SIMULTECH 2019: Proceedings of the 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications : Prague, Czech RepublicJuly 29-31, 2019}, editor = {Obaidat, Mohammad S. and {\"O}ren, Tuncer and Szczerbicka, Helena}, publisher = {INSTICC}, isbn = {978-989-758-381-0}, doi = {10.5220/0007932903170326}, pages = {317 -- 326}, year = {2019}, language = {en} } @inproceedings{BeckerLippelZielke2021, author = {Becker, Martin and Lippel, Jens and Zielke, Thomas}, title = {Dimensionality Reduction for Data Visualization and Linear Classification, and the Trade-off between Robustness and Classification Accuracy}, series = {Proceedings of ICPR 2020: 25th International Conference on Pattern Recognition, 10-15 January 2021, Milan}, booktitle = {Proceedings of ICPR 2020: 25th International Conference on Pattern Recognition, 10-15 January 2021, Milan}, editor = {Vezzani, Roberto}, publisher = {IEEE}, address = {Piscataway, NJ}, isbn = {978-1-72818-808-9}, doi = {10.1109/ICPR48806.2021.9412865}, pages = {6478 -- 6485}, year = {2021}, language = {en} } @inproceedings{BeckerLippelZielke2019, author = {Becker, Martin and Lippel, Jens and Zielke, Thomas}, title = {Gradient Descent Analysis: On Visualizing the Training of Deep Neural Networks}, series = {VISIGRAPP 2019: Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : Prague, Czech Republic, February 25-27, 2019}, volume = {Volume: 3}, booktitle = {VISIGRAPP 2019: Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : Prague, Czech Republic, February 25-27, 2019}, publisher = {INSTICC}, isbn = {978-989-758-354-4}, doi = {10.5220/0007583403380345}, pages = {338 -- 345}, year = {2019}, language = {en} } @inproceedings{StuhlsatzLippelZielke2010, author = {Stuhlsatz, Andr{\´e} and Lippel, Jens and Zielke, Thomas}, title = {Discriminative feature extraction with Deep Neural Networks}, series = {The 2010 International Joint Conference on Neural Networks (IJCNN), 18.07.2010 - 23.07.2010, Barcelona, Spain}, booktitle = {The 2010 International Joint Conference on Neural Networks (IJCNN), 18.07.2010 - 23.07.2010, Barcelona, Spain}, publisher = {IEEE}, isbn = {978-1-4244-6916-1}, doi = {10.1109/IJCNN.2010.5596329}, pages = {1 -- 8}, year = {2010}, language = {en} } @inproceedings{GoebelFrankZielkeetal.2017, author = {Goebel, Johannes and Frank, Lena and Zielke, Thomas and Becker, Martin and Lippel, Jens}, title = {Modellierung dynamischer Prozesse mit Deep Neural Networks am Beispiel einer Gas-Absorptionsw{\"a}rmepumpe}, series = {Deutsche Kaelte-und Klimatagung 2017 Bremen}, booktitle = {Deutsche Kaelte-und Klimatagung 2017 Bremen}, editor = {Deutscher K{\"a}lte- und Klimatechnischer Verein,}, publisher = {Curran Associates, Inc.}, address = {Red Hook}, pages = {1147 -- 1160}, year = {2017}, language = {de} } @article{StuhlsatzLippelZielke2012, author = {Stuhlsatz, Andr{\´e} and Lippel, Jens and Zielke, Thomas}, title = {Feature extraction with deep neural networks by a generalized discriminant analysis}, series = {IEEE transactions on neural networks and learning systems}, volume = {23}, journal = {IEEE transactions on neural networks and learning systems}, number = {4}, publisher = {IEEE}, doi = {10.1109/tnnls.2012.2183645}, pages = {596 -- 608}, year = {2012}, abstract = {We present an approach to feature extraction that is a generalization of the classical linear discriminant analysis (LDA) on the basis of deep neural networks (DNNs). As for LDA, discriminative features generated from independent Gaussian class conditionals are assumed. This modeling has the advantages that the intrinsic dimensionality of the feature space is bounded by the number of classes and that the optimal discriminant function is linear. Unfortunately, linear transformations are insufficient to extract optimal discriminative features from arbitrarily distributed raw measurements. The generalized discriminant analysis (GerDA) proposed in this paper uses nonlinear transformations that are learnt by DNNs in a semisupervised fashion. We show that the feature extraction based on our approach displays excellent performance on real-world recognition and detection tasks, such as handwritten digit recognition and face detection. In a series of experiments, we evaluate GerDA features with respect to dimensionality reduction, visualization, classification, and detection. Moreover, we show that GerDA DNNs can preprocess truly high-dimensional input data to low-dimensional representations that facilitate accurate predictions even if simple linear predictors or measures of similarity are used.}, language = {en} }