IMLA – Institute for Machine Learning and Analytics
Welcome to the IMLA
We conduct research on the intelligent analysis of large data sets and develop artificial intelligence applications. Through transfer projects, we apply our data expertise directly in the field. In addition, our knowledge is incorporated into the degree programs at Hochschule Offenburg, as well as into digital teaching materials and continuing education courses.
Research
The IMLA's primary mission is to carry out research and knowledge transfer projects on topics related to data analysis and artificial intelligence. The research questions addressed reflect the areas of expertise of the IMLA members.
Autonomous Systems
An autonomous system is a system that can make decisions on its own. To do this, it must perceive its environment, develop a model of that environment based on its perceptions, and act autonomously using that model. It makes decisions, which are then translated into planned actions. These actions, in turn, affect the environment. An autonomous system often interacts with other autonomous systems.
Examples include autonomous vehicles and robots that operate autonomously—systems that are not rigidly programmed but must react to changes in their environment. Projects involving autonomous systems at Hochschule Offenburg include the Audi Autonomous Driving Cup and the Shell Eco-Marathon Autonomous Challenge (autonomous driving), as well as Sweaty (robot soccer).
Big Data
In the field of big data, we investigate how very large volumes of structured and semi-structured data can be efficiently stored and, in particular, analyzed. A cluster running Hadoop and Spark is available for projects and proof-of-concepts, with data analysis performed using GPUs (on Hadoop) as well as R and Spark. The various components of Hadoop and Spark each have their own areas of application, but they can often be integrated with other tools or external systems to handle even more complex tasks. For example, the scope of such a platform can be expanded by incorporating machine learning tools.
As part of a project with an industrial company, for instance, a “big data platform” for storing and analyzing production data was designed and implemented. The use of Hadoop technologies complemented the relational database technologies currently in use and enabled faster (ad hoc) analyses as well as machine learning on the data.
Business Analytics
Business analytics focuses on the application of machine learning methods and statistical analyses to business management issues. These methods are applied across all areas of a company: from target audience analysis and campaign optimization in marketing, to sales forecasts, to the analysis of manufacturing processes based on sensor data. Traditional topics in data analysis and storage, such as business intelligence and data warehousing—as well as their extension through machine learning methods—are also the focus of our research interests. In the future, methods for processing natural language and other unstructured data will become increasingly important in business processes as well.
Learning Analytics
Learning analytics refers to the measurement, collection, analysis, and evaluation of data about learners and learning processes in order to provide support for learning and teaching using the collected data. A wide variety of machine learning methods and approaches are used in this process. The underlying data is typically generated in electronic learning environments and digital exam management systems. It is based, for example, on exam results, learning materials, online discussions, tests, and exercises. In all activities related to learning analytics, learners’ privacy and personal data must be protected.
Machine Learning
With the help of machine learning, computers are able to learn on their own, based on data, how to solve complex tasks without having to be explicitly programmed to do so. Machine learning algorithms use provided training and test data to search for patterns and learn a model that can be used to make decisions automatically in the future. The goal is for the computer to learn automatically and adapt its actions without human assistance.
There are a number of different machine learning methods, which can be categorized into supervised and unsupervised learning, as well as reinforcement learning. We work with all types of machine learning, with a particular focus on deep learning.
Multimedia Databases
Unlike traditional, structured databases, multimedia databases store unstructured data—primarily images, audio, and video, but also free-form text. A viewer can understand the content of an image or a video, but its meaning remains hidden from a computer. As a result, searching for multimedia content in a database is only truly feasible if semantic metadata is available to describe that content. Today, this metadata can be generated using deep learning; examples include objects in images and videos, or faces/people and their emotions. This would make it possible, for instance, to search for images, sounds, and videos featuring a tiger.
Parallel Computing
Parallel computing is a prerequisite for the practical implementation of many machine learning techniques. For example, parallelization reduces the training time for deep neural networks—which can sometimes take weeks—to just a few days or hours, making the practical application of deep learning possible in the first place. Many other machine learning methods are also highly computationally intensive and require parallel or distributed computing.
We focus in particular on massive parallelization using graphics processing units (GPUs) and algorithms optimized for them, e.g., in the field of data clustering. The IMLA has several servers equipped with NVIDIA Tesla GPUs for highly parallel data processing.
Smart Mobile Devices
Machine learning methods can also be used in a mobile context, particularly on smartphones. This can be achieved using various system architectures: On the one hand, the algorithms can be executed entirely on servers, meaning that the data to be analyzed must be transmitted from the smartphone via appropriate interfaces. However, this requires a constant connection to the servers. Alternatively, when using neural networks, for example, only the computationally intensive training can take place on the servers; the trained network is then transferred to the smartphone and used locally (see TensorFlow Lite). Another approach is to run machine learning applications entirely on the smartphone, for example, using hardware-accelerated SDKs such as ARCore (Android) or ARKit (iOS).
Research Projects
BMBF Project KI-Bohrer
Project website: www.ki-bohrer.de
Project Description
Geothermal energy as an alternative source of electricity and heat for achieving climate goals. Meeting internationally agreed-upon climate goals—including limiting global warming to below 1.5 degrees Celsius, if possible—is certainly one of the greatest technical and societal challenges of the coming decades. Without a further massive expansion of CO2-neutral energy generation, these goals will not be achievable. Geothermal power plants—particularly those located near urban areas and equipped with combined heat and power systems—could make a decisive contribution to climate-neutral heat and electricity supply in metropolitan areas. Provided a site is geologically suitable, geothermal plants offer several key advantages over other renewable technologies: I) Decentralized small- to medium-capacity plants enable local district heating supply and thus achieve very high efficiency; II) existing coal- and gas-fired power plants can be retrofitted sustainably and cost-effectively by utilizing the existing infrastructure; and III) when in operation, the plants are visually unobtrusive, quiet, and odorless, which contributes to a high level of public acceptance.
Problem: Noise Control for Urban Geothermal Drilling
In addition to the many operational advantages, however, the construction of geothermal plants in metropolitan areas also poses a significant technical and social problem: with currently available methods, the deep drilling required at urban sites leads to considerable and prolonged noise pollution for local residents. In particular, since drilling must continue continuously—24 hours a day, 7 days a week, for months on end—for technical and economic reasons, complying with legal noise limits of typically 35 dB (at night in residential areas) poses an enormous technical and logistical challenge. Current approaches to noise control rely on a combination of I) continuous monitoring of noise emissions through acoustic measurements in the surrounding area; II) optimization of construction site logistics, e.g., scheduling noisy activities for specific times of day; III) passive sound insulation through protective walls and enclosures around the drilling rigs; IV) active noise reduction through (currently) manual adjustment of drilling control parameters.
Approach: Active noise reduction through AI-based control of the drilling rigs
While Approaches I–III are currently part of the technical standard, we see considerable room for innovation and a need for further development in active noise reduction. The manual control methods currently in use are insufficient to address the comprehensive complexity of the optimization problem at hand: Noise control and the technically safe, economical advancement of the borehole are not only conflicting objectives; monitoring a wide range of operating conditions and simultaneously adjusting hundreds of control parameters can only be implemented suboptimally by manual means. Therefore, the goal of this project is to automate the optimization of drilling operations using state-of-the-art AI methods. We expect this fundamentally new approach to significantly improve noise protection for local residents while simultaneously increasing drilling speed. This is expected to significantly increase not only the public acceptance of geothermal projects but also their economic viability. Both of these improvements over the current state of the art could make a substantial contribution to the further expansion of geothermal energy and, consequently, to the achievement of overarching climate goals.
Q-AMeLiA
Quality Assurance of Machine Learning Applications (Q-AMeLiA)
The goal of the consortium comprising IMLA, Karlsruhe University of Applied Sciences, and Furtwangen University of Applied Sciences is to support SMEs in the specific machine learning software development life cycle (ML-SDLC) and the key quality indicators associated with it. Five SMEs are collaborating with three universities of applied sciences to develop appropriate tools for assessing data quality in terms of representative coverage of the feature space, as well as for evaluating the quality of the trained AI model achieved during the learning process. This mitigates the product risk for manufacturers of AI-based products and guarantees customers quantified product performance with regard to the AI’s decisions.
Project website: https://q-amelia.in.hs-furtwangen.de/
UPPER RHINE 4.0
On the topic of cross-border Industry 4.0 in the trinational Upper Rhine region, Hochschule Offenburg is participating in this joint competence network alongside numerous partners.
The main goal is to foster cross-border networking of Industry 4.0 expertise from northwestern Switzerland, the Grand-Est region of France, and Baden-Württemberg. This is intended to help the Upper Rhine region establish itself as a model for the successful implementation of Industry 4.0.
A wide range of activities takes place as part of the project, including workshops, continuing education programs, and student exchange activities. Cross-border events provide opportunities for direct exchange with industry and Universities in the region.
Among other things, the IMLA is organizing a cross-border summer school on machine learning.
The “UpperRhine 4.0” project is funded by the European Regional Development Fund (ERDF). The project runs from October 2017 to September 2020.
KompiLe
Project duration: December 1, 2021, to November 30, 2025
The “KompiLe” project, funded by the federal-state program “Artificial Intelligence in Higher Education,” examined the use of artificial intelligence (AI) in teaching.
A particular challenge posed by AI-supported technologies lies in balancing technical, ethical, social, and legal aspects and, on this basis, finding optimal solutions for higher education.
KompiLe directly linked learning with AI to learning about AI, thereby enabling a reflective, AI-supported learning process. The project was based on the assumption that the design and use of AI-based learning offerings require AI literacy on the part of both instructors and learners, while also fostering it.
About the Project
Learning with AI
To optimally support learning as an active, constructive, and individualized process, the KompiLe project developed an intelligent, adaptive learning environment based on learning preferences, experiences, and learning strategies.
Learning About AI
In addition, the following modules on AI-related content were designed, implemented, and evaluated.
AI in the Media
Chatbots
Artificial Intelligence—Ethics and Data Protection
Best practices for teaching these modules can be found here (PDF).
Publications
Journal articles, book chapters, and conference proceedings
Schmidt, C., Sedlmeier, T., Bauer, K., Canz, M., Schlemmer, D., & Sänger, V. (2025): Fostering AI Competence—Learning with and about Chatbots in a Making Scenario. Journal of Higher Education Development, ZFHE 20/Special Issue on Artificial Intelligence (double-blind peer review)
Schmidt, C. & Sänger, V. (2025): AI in Higher Education Teaching—Didactic Instruction and Learning Support. In: T. Breyer-Mayländer, D. Drechsler, C. Zerres (Eds.): AI Transformation in Germany. UTB, 2025.
Dahal, P., Nugroho, S., Schmidt, C., & Sänger, V. (2025): AI-Based Learning Recommendations: Use in Higher Education. Future Internet 2025, 17(7), 285. doi.org/10.3390/fi17070285
Schlemmer, D., Schmidt, C., Bauer, K., Canz, M., Sänger, V., & Sedlmeier, T. (2023). Promoting AI Competence: Pedagogical Making in Higher Education. Ludwigsburger Beiträge Zur Medienpädagogik, 23, 1–14. doi.org/10.21240/lbzm/23/11 (Double-Blind Peer Review)
Dahal, P., Nugroho, S., Schmidt, C., & Sänger, V. (2024). Practical Use of AI-Based Learning Recommendations in Higher Education. In: Herodotou, C., et al. Methodologies and Intelligent Systems for Technology-Enhanced Learning, 14th International Conference. MIS4TEL 2024. Lecture Notes in Networks and Systems, vol. 1171. Springer, Cham. doi.org/10.1007/978-3-031-73538-7_6
Sedlmeier, Teresa; Schmidt, Claudia; Sänger, Volker; Bauer, Katrin; Canz, Michael; Hillenbrand, Gisela; Dahal, Prabin; Nugroho, Saptadi (2024): Learning Experience through Content Curation and AI-Based Learning Recommendations. Proceedings of DELFI 2024. German Informatics Society (Gesellschaft für Informatik e.V.). ISSN: 2944-7682. EISSN: 2944-7682, DOI: doi.org/10.18420/delfi2024_32
Nugroho, S., Dahal, P., Hillenbrand, G., Bauer, K., Sedlmeier, T., Schlemmer, D., Schmidt, C., Sänger, V.: From LMS to LXP: Extending Moodle with AI-Based Recommendations for Learning. Upper Rhine Artificial Intelligence Symposium (URAI), September 17, 2023. Conference Proceedings: urai2023.sciencesconf.org/data/pages/book_urai2023_en_2024.pdf
Presentations
Sedlmeier, T. (2025): Toward a Learning Experience Platform with OER, Content Curation, and AI: A Functional Evolution of the Learning Management System. Keynote at the ORCA.nrw OER Symposium, September 11, 2025
Bauer, K., Canz, M.: KompiLe – Promoting AI Competence, Supporting Individualized Learning. Teaching Day, Hochschule Offenburg, April 18, 25
Sedlmeier, T., Schmidt, C., Sänger, V., Bauer, K., Canz, M., Hillenbrand, G., Dahal, P., & Nurgroho, S. (2024): Learning Experience Through Content Curation and AI-Based Learning Recommendations. 22nd Symposium on Educational Technologies (DELFI 2024), Fulda University of Applied Sciences, September 9–11, 24
Dahal, P., Nugroho, S., Schmidt, C., Sänger, V. (2024). Practical Use of AI-Based Learning Recommendations in Higher Education. 14th International Conference. MIS4TEL 2024, Salamanca, Spain. June 26–28, 2024
Hillenbrand, G: From LMS to LXP—Moodle Plugins for AI-Based Learning Recommendations, Moodle Conference, Leipzig, March 12–13, 2024
Nugroho, S., Dahal, P., Hillenbrand, G., Bauer, K., Sedlmeier, T., Schlemmer, D., Schmidt, C., Sänger, V.: From LMS to LXP: Extending Moodle with AI-Based Recommendations for Learning. Upper Rhine Artificial Intelligence Symposium (URAI), September 17, 2023
Sänger, V. & Schmidt, C.: The KompiLe Project: AI as a Learning Companion? Teaching Day, Hochschule Offenburg, April 27, 23
Sänger, V.: AI in Teaching—Ideas and Concepts from a Current Research Project, Freiburg. Baden Association of Industrial Enterprises (WvBI) Black Forest, March 8, 2024
Poster
Poster presentation (PDF) at the Teaching and Learning Conference: Teaching Innovations for Universities in the Digital World, Stuttgart, 2023
Poster presentation (PDF) at Teaching Day: Thematic Tables on AI Applications, Hochschule Offenburg, November 27, 2025
Project Management
Prof. Volker Sänger
Supported by
Completed Projects
TSAAI Project
TSAAI Project: Freely Available Online Continuing Education in Applied AI
After three years, the Transversal Skills in Applied AI (TSAAI) project will come to an end on February 28, 2025. The goal of the project was to develop an online course on applied AI for participants who do not have an explicit IT or Informatik background but come from a specific application domain.
Under the coordination of the University of Málaga, six European universities, including Hochschule Offenburg, have developed a freely available course.
Content: Introduction to AI and machine learning, AI methodology, applications of AI in industry and the IoT, the humanities, natural sciences, and the financial sector. With the exception of the foundational modules, participants can freely select and combine individual application areas as they wish. Example: Participants with a background in engineering choose the module “Applied AI in Industry and the IoT” (developed by TalTech in Tallinn, Estonia) with a focus on AI applications in the energy sector.
We have also attempted to incorporate current developments in generative AI, even though this poses a major challenge given the current pace and dynamics of the field.
Screenshot of the FuturIA learning platform showing the available modules
Format: Online, guided self-study. There are 8 modules in total, and the course is worth 15 credits. All course contents are freely available. A fee will apply for those wishing to obtain a certificate from the University of Málaga (currently in planning).
Target audience: Undergraduate students in all disciplines nearing the end of their bachelor’s degree, graduate students, and professionals who wish to deepen their knowledge of the fundamentals of AI and its various applications.
Languages: The material is available in English, German, Spanish, Slovenian, Croatian, Estonian, and Macedonian.
In the fall of 2024, the course underwent a successful initial pilot with students from the participating universities across various degree programs.
All information and access to the platform can be found at https://www.tsaai.eu/
Interested?
Please contact tobias.hagen@hs-offenburg.de, and we'd be happy to answer your questions in a one-on-one conversation!
Funding Notice
The project was funded by the EU as part of the Erasmus+ KA2 program.
MachineLearn-ING
Hochschule Offenburg has a growing continuing education sector that currently offers five master’s programs as well as certificate courses. The university is integrated into numerous regional networks, some of which are explicitly focused on digitalization and continuing education. These networks are being utilized for the MachineLearn-ING pilot project to encourage working high-level professionals to pursue continuing education. The technical focus of MachineLearn-ING is machine learning, which is considered a key driver of innovation for digitalization. The organizational structure of the continuing education course is tailored to the target group’s tight schedule.
The project thus serves as a model case for reaching specific target groups with the future-oriented qualification programs of the “Workplace 4.0.”
The project is funded through the SmartQualifiziert program, which is part of the Stifterverband’s Future Skills initiative.
Predictive Maintenance
The “Predictive Maintenance” research project is receiving 750,000 euros in funding from the Carl Zeiss Foundation’s “Transfer” funding program. The project will begin on January 1, 2019, and the funding period is three years. The research project focuses on the development of an Industry 4.0-compatible technology for the functional and procedural design of predictive and intelligent maintenance solutions.
ML2
The interdisciplinary research project “Menschen lernen Maschinelles Lernen” (ML2) addressed the question: How can we harness the potential of machine learning for small and medium-sized enterprises while ensuring that students receive application-oriented training?
It was funded by the Federal Ministry of Education and Research (BMBF) as part of the “IKT 2020 – Research for Innovation” funding program and implemented by the Analytics and Data Science research group at Hochschule Offenburg. The project provided training in machine learning to both company employees and students. The curriculum designed for this purpose included not only theoretical content but also practical components in which participants had to solve current problems at the participating companies using machine learning. A total of approximately 40 employees and an equal number of students completed the curriculum.
Insights into Our Research
Developing new methods. Optimizing processes. Driving innovation. At IMLA, we seek answers to research questions. Our project directory lists all the projects we’re carrying out in collaboration with partners from academia and industry. There, you can search for all ongoing and completed projects since 2014. You can find the latest milestones and breakthroughs in our daily work under “Insights.”
Industrial Partnerships
Companies can collaborate with the IMLA in various ways and thus benefit from our expertise in research and technology transfer:
As a project partner in a publicly funded research or technology transfer project. Depending on the arrangement, the company may receive funding itself or participate as a non-funded partner to share in the project results.
As a client commissioning a project. In this case, you retain all rights to the results. A project may involve just a few person-days or, for example, be carried out as part of a doctoral dissertation.
You have a research question that can be addressed as part of a student’s final thesis and supervised by a member of the IMLA on behalf of the university.
Long-term cooperation agreements with the University that go beyond individual projects are also possible. Please contact us!
Our Industry Partners
Education and Continuing Education
In addition to research and knowledge transfer, teaching and continuing education are also among the IMLA’s responsibilities. Data literacy is one of the key competencies of the information society. As a result, a large portion of the degree programs at Hochschule Offenburg include content on analytics and AI in their curricula, which is mostly taught by IMLA members. The spectrum ranges from introductory courses for engineering and Betriebswirtschaft programs to the specialized bachelor's program in Angewandte Künstliche Intelligenz.
Research and teaching on AI converge in the robot soccer projects: Sweaty and magmaOffenburg regularly achieve success at world and European championships, not least thanks to the great dedication of the participating students.
In addition, the IMLA develops teaching materials on analytics and AI, such as the online course on machine learning with practical exercises based on the KNIME platform, or as part of the Erasmus+-funded project “Transversal Skills in Applied Artificial Intelligence” (TSAAI).
Further Information
Publications
List of all recent IMLA publications on Google Scholar: https://scholar.google.de/citations?hl=de&user=XgNf1vYAAAAJ
Title |
| Year |
Fake or JPEG? Revealing Common Biases in Generated Image Detection Datasets |
| 2024 |
Urban Sound Propagation: A Benchmark for 1-Step Generative Modeling of Complex Physical Systems |
| 2024 |
Are Vision Language Models Texture or Shape Biased and Can We Steer Them? |
| 2024 |
Challenging the Black Box: A Comprehensive Evaluation of Attribution Maps of CNN Applications in Agriculture and Forestry |
| 2024 |
An in-depth study of U-net for seismic data conditioning: Multiple removal by moveout discrimination |
| 2024 |
Retail-786k: A Large-Scale Dataset for Visual Entity Matching |
| 2023 |
Improving Native CNN Robustness with Filter Frequency Regularization |
| 2023 |
Don't Look into the Sun: Adversarial Solarization Attacks on Image Classifiers |
| 2023 |
Deep Diffusion Models for Seismic Processing |
| 2023 |
Fix Your Downsampling ASAP! Be Natively More Robust via Aliasing- and Spectral Artifact-Free Pooling |
| 2023 |
As large as it gets: Learning infinitely large filters via neural implicit functions in the Fourier domain |
| 2023 |
Automating Wood Species Detection and Classification in Microscopic Images of Fibrous Materials with Deep Learning |
| 2023 |
On Invariance, Equivariance, Correlation, and Convolution of Spherical Harmonic Representations for Scalar and Vectorial Data |
| 2023 |
Seismic Demultiplexing with Deep Learning |
| 2023 |
Deep Diffusion Models for Multiple Removal |
| 2023 |
Fine-Grained Product Classification on Leaflet Advertisements |
| 2023 |
Deep Learning Strategies for Seismic Demultiple |
| 2023 |
The Power of Linear Combinations: Learning with Random Convolutions |
| 2023 |
On the Interplay of Convolutional Padding and Adversarial Robustness |
| 2023 |
Detecting Images Generated by Deep Diffusion Models Using Their Local Intrinsic Dimensionality |
| 2023 |
An Extended Study of Human-Like Behavior Under Adversarial Training |
| 2023 |
Rethinking 1×1 Convolutions: Can We Train CNNs with Frozen Random Filters? |
| 2023 |
Unfolding local growth rate estimates for (almost) perfect adversarial detection |
| 2022 |
Robust models are less overconfident |
| 2022 |
Physics-Constrained Deep Learning of Aerosol Microphysics |
| 2022 |
Aliasing and Adversarial Robust Generalization of CNNs |
| 2022 |
Does Medical Imaging Learn Different Convolution Filters? |
| 2022 |
Frequency-Low-Cut Pooling—Plug-and-Play Against Catastrophic Overfitting |
| 2022 |
GSparsity: Unifying Network Pruning and Neural Architecture Search by Group Sparsity |
| 2022 |
Dissecting U-Net for Seismic Applications: An In-Depth Study on Deep Learning for Multiple Removal |
| 2022 |
Dissecting U-Net for Seismic Applications: An In-Depth Study on Deep Learning Multiple Removal |
| 2022 |
Image-to-Image Seismic Interpolation |
| 2022 |
Intelligent Data Governance and Data Management—New Opportunities for Customer Data Management |
| 2022 |
New Channels—New Data: The Changing Role of Customer Data in Retail |
| 2022 |
An Empirical Investigation of Model-to-Model Distribution Shifts in Trained Convolutional Filters |
| 2022 |
[AutoMLConf'22]: GSparsity: Unifying Network Pruning and Neural Architecture Search by Group Teaser |
| 2022 |
An Empirical Investigation of Trained Convolutional Filters |
| 2022 |
Tackling Key Challenges of AI Development—Insights from an Industry-Academia Collaboration |
| 2022 |
Physics-Informed Learning of Aerosol Microphysics |
| 2022 |
Adversarial Robustness Through the Lens of Convolutional Filters |
| 2022 |
CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters |
| 2022 |
Investigating Shifts in GAN Output Distributions |
| 2021 |
Aliasing Coincides with CNNs’ Vulnerability to Adversarial Attacks |
| 2021 |
Is RobustBench/AutoAttack a suitable benchmark for adversarial robustness? |
| 2021 |
Detecting AutoAttack perturbations in the frequency domain |
| 2021 |
FacialGAN: Style transfer and attribute manipulation on synthetic faces |
| 2021 |
Emulating aerosol microphysics with machine learning |
| 2021 |
Spectraldefense: Detecting adversarial attacks on CNNs in the Fourier domain |
| 2021 |
Generative Models for the Transfer of Knowledge in Seismic Interpretation with Deep Learning |
| 2021 |
Estimating the Robustness of Classification Models Based on the Structure of the Learned Feature Space |
| 2021 |
A Retail Product Categorization Dataset |
| 2021 |
Local Facial Attribute Transfer through Inpainting |
| 2021 |
Learning embeddings for image clustering: An empirical study of triplet loss approaches |
| 2021 |
Investigating Shifts in GAN Output Distributions |
| 2021 |
KOBRA: Practical Learning-Based Methods for the Automatic Configuration of Business Rules in Duplicate Detection Systems |
| 2021 |
AI Projects—The Role of Data Quality |
| 2021 |
Checkout Optimization |
| 2021 |
Valid Customer Data—The Foundation for Omni-Channel Marketing |
| 2021 |
A Question of Quality |
| 2021 |
Sample-Efficient Localization and Stage Prediction with Autoencoders. |
| 2021 |
MSM: Multi-stage multicuts for scalable image clustering |
| 2021 |
Combating Mode Collapse in GAN Training: An Empirical Analysis Using Hessian Eigenvalues |
| 2021 |
Combining Transformer Generators with Convolutional Discriminators |
| 2021 |
Group Sparsity: A Unified Framework for Network Pruning and Neural Architecture Search |
| 2021 |
Combining Transformer Generators with Convolutional Discriminators |
| 2021 |
Latent Space Conditioning on Generative Adversarial Networks |
| 2021 |
Combating mode collapse in GAN training: An empirical analysis using Hessian eigenvalues |
| 2020 |
Latent Space Conditioning on Generative Adversarial Networks |
| 2020 |
An Empirical Study of Explainable AI Techniques on Deep Learning Models for Time Series Tasks |
| 2020 |
Python Workflows on HPC Systems |
| 2020 |
Teaching Practical Machine Learning Concepts to Professionals and Students: An Integrated and Interdisciplinary Qualification Project |
| 2020 |
Synthesizing Seismic Diffractions Using a Generative Adversarial Network |
| 2020 |
System setup for synchronized visual-inertial localization and mapping |
| 2020 |
Nothing Is as It Seems: On the Art of Collective Memory |
| 2020 |
Technique for monitoring technical equipment |
| 2020 |
Extracting Horizon Surfaces from 3D Seismic Data Using Deep Learning |
| 2020 |
Toward visual debugging for multi-target time series classification |
| 2020 |
PHS: A toolbox for parallel hyperparameter search |
| 2020 |
Detection of point scatterers using diffraction imaging and deep learning |
| 2020 |
Methodology for mapping a processing area for autonomous robot vehicles |
| 2020 |
Unsupervised multiple person tracking using autoencoder-based lifted multicuts |
| 2020 |
Message from the MLHPC Workshop Chairs |
| 2020 |
SmartPred: Unsupervised hard disk failure detection |
| 2020 |
Watch your up-convolution: CNN-based generative deep neural networks are failing to reproduce spectral distributions |
| 2020 |
Unsupervised Multiple Person Tracking Using Autoencoder-Based Lifted Multicuts |
| 2020 |
Python Workflows on HPC Systems |
| 2020 |
Synthesizing Seismic Diffractions Using a Generative Adversarial Network |
| 2020 |
Protect Our Health with Cleaner Cars—How to Gain Customer Acceptance for the Purchase |
| 2020 |
Learning to Walk with Toes |
| 2020 |
Experimental Setup for the Evaluation of Algorithms for Simultaneous Localization and Mapping |
| 2020 |
Machine Learning Meets Visualization to Make Artificial Intelligence Interpretable (Dagstuhl Seminar 19452) |
| 2020 |
Collaborations Between Industry and University |
| 2020 |
A Two-Stage Minimum-Cost Multicut Approach to Self-Supervised Multiple Person Tracking |
| 2020 |
Semi-few-shot attribute translation |
| 2019 |
Scalable hyperparameter optimization with lazy Gaussian processes |
| 2019 |
Gradvis: Visualization and second-order analysis of optimization surfaces during the training of deep neural networks |
| 2019 |
Autonomous work device |
| 2019 |
Unmasking deepfakes with simple features |
| 2019 |
Architecture of a Big Data Platform for a Semiconductor Company |
| 2019 |
Image-based automated hit detection and score calculation on a steel dartboard |
| 2019 |
Team
Management & Contact