For full functionality of this site it is necessary to enable JavaScript. Here are the instructions how to enable JavaScript in your web browser.

Position within the page tree

Home
Teaching
Teaching overview

Teaching Overview

Institute for Artificial Intelligence (KI)

Check bellow our course offerings over the semester.

In the following, you can find the teaching activities organized by each Department of the Insitute of Artificial Intelligence.

Please always use the links available on C@mpus if you are interested in enrolling.

Analytic Computing

Probabilistic Machine Learning

Lecture

Number:	023800600
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab
Type:	Lecture
Semester hours:	3
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	• Probabilistische Graphische Modelle (Bayes’sche Netze, Markov Networks, HMMs, Conditional Random Fields) • Inferenzmethoden (Variable Elimination, Message Passing) • Lernmethoden (Parameterlernen, Bayes’sche Methoden, Struktur-lernen) • Erklärbarkeit gelernter Modelle und Lernen mit Erklärungen (Lime, Shap Values, Explanation-based Learning...) • Structure and Causal Inference (Causal Bayes’ian Networks, Inter-ventions, Counterfactuals)
Link:	C@MPUS

Probabilistic Machine Learning

Exercise

Number:	023800605
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab
Type:	Exercise
Semester hours:	1
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	siehe Vorlesung
Link:	C@MPUS

Oberseminar Artificial Intelligence

Colloquium

Number:	021898001
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab
Type:	Colloquium
Semester hours:	2
Teaching Language:	German/English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	nur für Mitarbeiterinnen und Doktorandinnen
Link:	C@MPUS

Machine Learning with Graphs

Advanced Seminar

Number:	023895000
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Daniel Hernandez Cosimo Gregucci
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	Target Group Participants are expected to have completed the course “Machine Learning” or a related course in the field of machine learning. Familiarity with knowledge graphs is a plus. The Topic Nowadays, applications from multiple domains use graphs to store their data. Graph databases have been shown to be suitable for storing data of citation networks, proteins, social networks, circular economies, and general knowledge. Interesting problems like fraud detection and driver intention prediction can be formulated as graph machine learning problems like entity classification and link prediction, and then solved with techniques like graph neural networks, graph embeddings, and transformers. In this English-speaking Hauptseminar (presentations and seminar reports will have to be delivered in English) we will survey current approaches for applying machine learning on different graphs for different tasks. Process After an introduction on the topic, participants will work on a technical report about a chosen topic. The report will be peer-reviewed by other participants. Students will give a talk about their topic and hand in their technical report.
Link:	C@MPUS

Laboratory Course Artificial Intelligence: Deep Learning Lab

Exercise

Number:	023887001
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Mojtaba Nayyeri Lluís Pastor Pérez
Type:	Exercise
Semester hours:	4
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	The Topic This lab aims to teach the best available practices to solve problems of the main machine learning Tasks. This includes, but is not limited to, natural language processing and graph analysis, using the deep learning method. We provide various real-world examples with sufficient training data. Students can use these test cases to implement what they learn in the lab. Participants will select one of the provided topics, and build the prediction models using deep learning libraries (e.g. Pytorch). Students are expected to work in groups and learn the fundamental steps in developing accurate (thus effective) deep neural networks for predictions. This includes improving the overall pipelines through dataset fetching optimizations, data pre-processing, and enhancing model predictions through customizing the network architectures. Furthermore, students are encouraged to develop their ideas and practices to further improve the quality of their code. At the end of the course, students are expected to prepare a scientific report and give a presentation of the developed models, where a discussion with their peers would be established. Procedure At the beginning of the lab, the structure of their software codes, presentations, and reports will be discussed with the lecturers. Participants then report and discuss their progress weekly. At the end of the Lab, students give their final presentations. Hence, the reports must be handed in before the final presentations and the end of the summer term
Link:	C@MPUS

Foundation Models

Lecture

Number:	023800400
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Univ.-Prof. Ph.D. Mathias Niepert Dr. rer. nat. Amin Totounferoush
Type:	Lecture
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	Foundation Models, such as GPT4 and Stable Diffusion, are models that are trained on very large amounts of data and can be effectively fine-tuned for a wide range of downstream tasks. This lecture aims to convey the key ideas behind the recent Large Language Models and Multimodal Foundation Models. The students will learn both, fundamental building blocks and optimization techniques, as well as advanced architectures and recent developments in the field of Foundation Models. Among others, the following topics will be covered: • Fundamentals: overview of the relevant building blocks and optimization techniques (e.g., transformer models, self-supervised learning), generative models. • Foundation models for text processing • Vision-based and multimodal foundation models • Discussion of ethical, societal, and legal facets of FMs Alongside the lectures, student projects are an integral part of this lecture. These projects focus on analysing and presenting recent research papers as well as implementing/using them for a specific use-case or application to gain hands-on experience. The students will present their results and write a report summarizing their learnings from the project.
Link:	C@MPUS

Foundation Models

Exercise

Number:	023800405
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Univ.-Prof. Ph.D. Mathias Niepert Dr. rer. nat. Amin Totounferoush
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	siehe Vorlesung
Link:	C@MPUS

Formal Methods for the Web

Advanced Seminar

Number:	023896000
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Daniel Hernandez
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Analytic Computing
Content:	Target Group We expect participants to have completed courses covering basic topics of logic, algebra, set theory, and computer language theory. Experience on the Web, like the knowledge graphs course, is a plus. The Topic Initially, the Web was limited to a few scientific and technical areas. It is now universal. All information is moving online. Much of it was unstructured text data. Today, websites and web applications are recasting the data as semistructured XML, JSON, and RDF files. Diverse database systems interact in an ecosystem that integrates multiple data sources and services, requiring the definition of multiple standards. The Web is not only about data but about communication protocols, people, and fairness. The W3C and community groups are constantly discussing and proposing new standards for the Web. These proposals are usually ambiguous collections of descriptions and use case examples. Vendors implement standards differently because of the non-agreement of what the specifications say. Formal methods have been shown to be the way to make standards understandable and accessible to developers and scientists. In this course, you will learn and practice the foundations of the formal models for the Web and how to describe diverse types of elements, like computer languages, people's interactions in social networks, documents, data, and communication models. Process There will be regular lectures presenting formal methods for the Web, which will be alternated with student work. Students will read selected example papers formalizing Web standards, apply formal methods to unformalized Web standards, give a talk, and write a report (in English) summarizing their work. The report will be peer-reviewed by other participants.
Link:	C@MPUS

Socially Intelligent Robotics Lab (Autonomous Systems)

Oberseminar Intelligent Robotics

Colloquium
2 Sh
WS 25/26

Number:	024000110
Lecturers:	Univ.-Prof. Dr. Kai Arras Till Hielscher Dennis Rotondi Fabio Scaparro Abdelrhman Werby
Type:	Colloquium
Semester hours:	2
Teaching Language:	German
Semester:	WS 25/26
Organisation:	Autonome Systeme
Content:	In this Oberseminar, we review and discuss recent research trends in interactive and intelligent systems that perceive their surrounding with sensors, interact with humans and act intelligently in complex environments. We will accommodate talks on interesting papers, intermediate student project presentations, invited talks, tutorials, PhD student presentations on their recent research and results, summaries of conference visits, and so on. The topics include - Multimodal perception, tracking and analysis of humans - 3D Semantic SLAM and environment learning - Open-vocabulary environment and human activity modeling - Social and physical human-robot interaction - Imitation learning - Human affect recognition - Robot design
Link:	C@MPUS

Mobile Robotics

Lecture
2 Sh
WS 25/26

Number:	024000100
Lecturers:	Univ.-Prof. Dr. Kai Arras
Type:	Lecture
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Autonome Systeme
Content:	Probabilistic foundations, sensors for mobile robots, error propagation, filtering (Bayes filter, Kalman filter, particle filter), tracking and data association, localization method (odometry, Markov and Monte Carlo localization, EKF localization), grid mapping, SLAM (EKF SLAM, graph-based SLAM), search and motion planning for wheeled robots
Link:	C@MPUS

Mobile Robotics

Exercise
2 Sh
WS 25/26

Number:	024000105
Lecturers:	Univ.-Prof. Dr. Kai Arras Till Hielscher Dennis Rotondi Fabio Scaparro Abdelrhman Werby
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Autonome Systeme
Content:	siehe Vorlesung
Link:	C@MPUS

Advanced seminar “Research Topics for Intelligent Interactive Robots”

Advanced Seminar
2 Sh
WS 25/26

Number:	024000300
Lecturers:	Univ.-Prof. Dr. Kai Arras Till Hielscher Dennis Rotondi Fabio Scaparro Abdelrhman Werby
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Autonome Systeme
Content:	Advanced Seminar “Research Topics for Intelligent Interactive Robots” Interactive robots that perceive their surroundings, act intelligently and collaborate with number of application domains. Underlying research questions include, for example, learning, sensory analysis of humans, imitation learning, human behavior modeling a manipulation, and social human-robot interaction. Learning to efficiently and effectively read a paper is a critical skill taught and practice selection of recent, particularly relevant scientific publications from which you will pick read, analyze, summarize, present the paper, and in addition, implement it, or run experiments. We will deal with questions such as: 1. What is the underlying research question addressed in the paper? 2. Why is it relevant? 3. How does the paper relate to other papers? 4. What are the paper’s contributions and hypotheses? 5. How are the experiments designed to investigate the hypotheses? 6. Can you reproduce the results? 7. How does the paper’s method work under conditions different from the reported ones? 8. What are the flaws and limitations of the paper and its proposed method? 9. How would you extend the work? During the seminar, students are expected to contribute with questions, answers, comments, and observations. We emphasize the importance of everyone attending class and actively engaging in all discussions. Requirements/Tasks • Read and understand your paper thoroughly to answer questions 1-5 (don’t underestimate the time necessary for that) • Implement the paper or run it’s code to answer questions 6-9 • Present the paper and your experimental result in two talks of around 15 min plus 5 min for discussion • Attend the seminar and actively participate in the discussions. Read all presented papers in "first-pass" fashion (you will learn how to do this). • Talks can either be given in German or English (we recommend English) • The final mark is a combination of three grades for presentation, implementation, and active participation
Link:	C@MPUS

Machine Learning for Simulation Science

Theoretical and Methodological Foundations of Autonomous Systems

Lecture
2 Sh
WS 23/24

Number:	021821000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	2
Teaching Language:	English
Semester:	WS 23/24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	for M.Sc. Computer Science students: You have to attend the course "foundations of artificial intelligence" (LV-no. 020411000) https://campus.uni-stuttgart.de/cusonline/pl/ui/$ctx/wbLv.wbShowLVDetail?pStpSpNr=302564&pSpracheNr=1 If you registere here, we will transfer your registration to the above mentioned course as this is actually the course this module refers to.
Link:	C@MPUS

Theoretical and Methodological Foundations of Autonomous Systems

Exercise
2 Sh
WS 23/24

Number:	021822000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	WS 23/24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	for M.Sc. Computer Science students: You have to attend the course "foundations of artificial intelligence" (LV-no. 020411000) https://campus.uni-stuttgart.de/cusonline/pl/ui/$ctx/wbLv.wbShowLVDetail?pStpSpNr=302564&pSpracheNr=1 If you registere here, we will transfer your registration to the above mentioned course as this is actually the course this module refers to.
Link:	C@MPUS

Theoretical and Methodological Foundations of Autonomous Systems

Exercise
2 Sh
WS 24/25

Number:	021822000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	WS 24/25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	for M.Sc. Computer Science students: You have to attend the course "foundations of artificial intelligence" (LV-no. 020411000) https://campus.uni-stuttgart.de/cusonline/pl/ui/$ctx/wbLv.wbShowLVDetail?pStpSpNr=302564&pSpracheNr=1 If you registere here, we will transfer your registration to the above mentioned course as this is actually the course this module refers to.
Link:	C@MPUS

Theoretical and Methodological Foundations of Autonomous Systems

Lecture
2 Sh
WS 24/25

Number:	021821000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	2
Teaching Language:	English
Semester:	WS 24/25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	for M.Sc. Computer Science students: You have to attend the course "foundations of artificial intelligence" (LV-no. 020411000) https://campus.uni-stuttgart.de/cusonline/pl/ui/$ctx/wbLv.wbShowLVDetail?pStpSpNr=302564&pSpracheNr=1 If you registere here, we will transfer your registration to the above mentioned course as this is actually the course this module refers to.
Link:	C@MPUS

Reinforcement Learning

Exercise
2 Sh
SS 24

Number:	021862000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	SS 24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Siehe Vorlesung
Link:	C@MPUS

Reinforcement Learning

Exercise
2 Sh
SS 25

Number:	021862000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	SS 25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Siehe Vorlesung
Link:	C@MPUS

Reinforcement Learning

Lecture
2 Sh
SS 25

Number:	021861000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	2
Teaching Language:	English
Semester:	SS 25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Reinforcement Learning considers how an agent, interacting with a world, can improve or learn optimal behavior based on own experience or teacher demonstration. This branch of Artificial Intelligence and Machine Learning has become increasingly important foundation of robust intelligent systems and robotics. Optimal exploration (behavior that optimizes the agent's information gain) is a particularly interesting aspect of Reinforcement Learning. This lecture will introduce to the theory of Reinforcement Learning and then discuss state-of-the-art algorithms in this area. A focus of the lecture will be on deep reinforcement learning. Markov Decision Processes and Bellman's optimality principle basic model-free RL methods (policy gradient, Q-learning, etc) model-based RL methods offline reinforcement learning relational RL inverse RL, learning from demonstration and instruction basics of reinforcement learning theory transfer and multi-task learning applications
Link:	C@MPUS

Reinforcement Learning

Lecture
2 Sh
SS 24

Number:	021861000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	2
Teaching Language:	English
Semester:	SS 24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Reinforcement Learning considers how an agent, interacting with a world, can improve or learn optimal behavior based on own experience or teacher demonstration. This branch of Artificial Intelligence and Machine Learning has become increasingly important foundation of robust intelligent systems and robotics. Optimal exploration (behavior that optimizes the agent's information gain) is a particularly interesting aspect of Reinforcement Learning. This lecture will introduce to the theory of Reinforcement Learning and then discuss state-of-the-art algorithms in this area. A focus of the lecture will be on deep reinforcement learning. Markov Decision Processes and Bellman's optimality principle basic model-free RL methods (policy gradient, Q-learning, etc) model-based RL methods offline reinforcement learning relational RL inverse RL, learning from demonstration and instruction basics of reinforcement learning theory transfer and multi-task learning applications
Link:	C@MPUS

Recent Applications of Machine Learning

Seminar
2 Sh
SS 25

Number:	021890000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Ho Minh Duy Nguyen
Type:	Seminar
Semester hours:	2
Teaching Language:	English
Semester:	SS 25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	https://www.f05.uni-stuttgart.de/informatik/studierende/seminaranmeldung/
Link:	C@MPUS

Recent Applications of Machine Learning

Seminar
2 Sh
SS 24

Number:	021890000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Ho Minh Duy Nguyen
Type:	Seminar
Semester hours:	2
Teaching Language:	English
Semester:	SS 24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	https://www.f05.uni-stuttgart.de/informatik/studierende/seminaranmeldung/
Link:	C@MPUS

Machine Learning for Simulation Science

Colloquium
2 Sh
WS 25/26

Number:	021898003
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Colloquium
Semester hours:	2
Teaching Language:	German/English
Semester:	WS 25/26
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	nur für Mitarbeiterinnen und Doktorandinnen
Link:	C@MPUS

Machine Learning for Simulation Science

Colloquium
2 Sh
WS 23/24

Number:	021898003
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Colloquium
Semester hours:	2
Teaching Language:	German/English
Semester:	WS 23/24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	nur für Mitarbeiterinnen und Doktorandinnen
Link:	C@MPUS

Machine Learning for Simulation Science

Colloquium
2 Sh
WS 24/25

Number:	021898003
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Colloquium
Semester hours:	2
Teaching Language:	German/English
Semester:	WS 24/25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	nur für Mitarbeiterinnen und Doktorandinnen
Link:	C@MPUS

Kolloquium Machine Learning for Simulation Science

Colloquium
2 Sh
SS 24

Number:	021898000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Colloquium
Semester hours:	2
Teaching Language:	German
Semester:	SS 24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	nur für Mitarbeiter*innen und PhD-Studierende
Link:	C@MPUS

Kolloquium Machine Learning for Simulation Science

Colloquium
2 Sh
SS 25

Number:	021898000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Colloquium
Semester hours:	2
Teaching Language:	German
Semester:	SS 25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	nur für Mitarbeiter*innen und PhD-Studierende
Link:	C@MPUS

Grundlagen der Künstlichen Intelligenz

Exercise
1 Sh
WS 24/25

Number:	020412000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	1
Teaching Language:	English/German
Semester:	WS 24/25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	siehe Vorlesung
Link:	C@MPUS

Grundlagen der Künstlichen Intelligenz

Lecture
3 Sh
WS 24/25

Number:	020411000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	3
Teaching Language:	English
Semester:	WS 24/25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Intelligenz Agentenbegriff Problemlösen durch Suchen, Suchverfahren Probleme mit Rand- und Nebenbedingungen Spiele Aussagen- und Prädikatenlogik Logikbasierte Agenten, Wissensrepräsentation Inferenz Planen Unsicherheit, probabilistisches Schließen Probabilistisches Schließen über die Zeit Entscheidungstheorie
Link:	C@MPUS

Grundlagen der Künstlichen Intelligenz

Exercise
1 Sh
WS 23/24

Number:	020412000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	1
Teaching Language:	English/German
Semester:	WS 23/24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	siehe Vorlesung
Link:	C@MPUS

Grundlagen der Künstlichen Intelligenz

Lecture
3 Sh
WS 23/24

Number:	020411000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	3
Teaching Language:	English
Semester:	WS 23/24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Intelligenz Agentenbegriff Problemlösen durch Suchen, Suchverfahren Probleme mit Rand- und Nebenbedingungen Spiele Aussagen- und Prädikatenlogik Logikbasierte Agenten, Wissensrepräsentation Inferenz Planen Unsicherheit, probabilistisches Schließen Probabilistisches Schließen über die Zeit Entscheidungstheorie
Link:	C@MPUS

Foundations of Artificial Intelligence / Theoretical and Methodological Foundations of Autonomous Systems

Lecture
3 Sh
WS 25/26

Number:	020411000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	3
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Classical Search, Local Search, Constraint Satisfaction Problems, Adversarial Search and Games, Knowledge-based agents based on logic, uncertainty in artificial intelligence, probabilistic representations and reasoning, rational decision-making, Markov decision processes, brief introduction to machine learning, and deep learning. Selective topics such as generative AI and ethical implications of AI systems.
Link:	C@MPUS

Foundations of Artificial Intelligence / Theoretical and Methodological Foundations of Autonomous Systems

Exercise
1 Sh
WS 25/26

Number:	020412000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	1
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	siehe Vorlesung
Link:	C@MPUS

Deep Learning for the Sciences

Advanced Seminar
2 Sh
WS 23/24

Number:	021895000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Marimuthu Kalimuthu
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	WS 23/24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Deep learning has been successfully applied in a wide range of use cases and specifically in applications involving visual and textual data. Modern machine translation systems and search engines, for example, are using language models trained on large text corpora. Increasingly, deep learning is also applied to problems arising in the sciences and engineering. For instance, deep learning for graphs is used to learn simulators from data. The figure on the right shows a particle simulation obtained from a graph neural network trained on simulation traces from a numerical solver. The advantage of using deep learning in this context is its ability to integrate the simulation into a larger neural network with a corresponding loss function and train the resulting model end-to-end. Other applications of deep learning can be found in chemistry, the biomedical sciences, and engineering disciplines, where use cases range from drug-protein interaction prediction to modeling fluid dynamics. Since machine learning and specifically deep learning will be increasingly used in disciplines of science and engineering, this seminar’s goal is to provide an overview of applications, to give students a deeper understanding of recent work, and to have an opportunity to learn reading, analyzing, and engaging with scientific papers. If possible, the students are also encouraged to demo the application to the seminar.
Link:	C@MPUS

Deep Learning for the Sciences

Advanced Seminar
2 Sh
SS 24

Number:	021895000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Marimuthu Kalimuthu
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	SS 24
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	https://www.f05.uni-stuttgart.de/informatik/studierende/seminaranmeldung/
Link:	C@MPUS

Deep Learning for the Sciences

Advanced Seminar
2 Sh
WS 24/25

Number:	021895000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Daniel Musekamp Edgar Torres Francesca Quaranta
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	WS 24/25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Link:	C@MPUS

Deep Learning for the Sciences

Advanced Seminar
2 Sh
WS 25/26

Number:	021895000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Daniel Musekamp Edgar Torres
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	WS 25/26
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	Deep learning has been successfully applied in a wide range of use cases and specifically in applications involving visual and textual data. Modern machine translation systems and search engines, for example, are using language models trained on large text corpora. Increasingly, deep learning is also applied to problems arising in the sciences and engineering. For instance, deep learning for graphs is used to learn simulators from data. The figure on the right shows a particle simulation obtained from a graph neural network trained on simulation traces from a numerical solver. The advantage of using deep learning in this context is its ability to integrate the simulation into a larger neural network with a corresponding loss function and train the resulting model end-to-end. Other applications of deep learning can be found in chemistry, the biomedical sciences, drug discovery, and engineering disciplines, where use cases range from drug-protein interaction prediction to modeling fluid dynamics. Since machine learning and specifically deep learning will be increasingly used in disciplines of science and engineering, this seminar’s goal is to provide an overview of applications, to give students a deeper understanding of recent work, and to have an opportunity to learn how to read, analyze, and engage with scientific papers. Target group Participants of the seminar are expected to have completed the course “Machine Learning”, “Reinforcement Learning”, and/or related courses. Organization Students of the English-speaking seminar are encouraged to select their own presentation topic, typically based on 1-2 papers and/or a blog post. Alternatively, students can choose from a list of pre-selected topics. In addition to presenting the concrete application of DL, the student will give a brief introduction to the deep learning method used in the paper. Students will give English presentations in the weekly seminar. 2 weeks before the presentation, the structure of their presentations will be discussed with the lecturers. After the presentations and before the end of the summer term, the seminar reports must be handed in. Reports are written as blog posts and if the students agree, can be published. The seminar is organized by Mathias Niepert who has extensive experience developing (deep) machine learning methods and applying these methods to problems from the biomedical and medical sciences.
Link:	C@MPUS

Deep Learning for the Sciences

Advanced Seminar
2 Sh
SS 25

Number:	021895000
Lecturers:	Univ.-Prof. Ph.D. Mathias Niepert Marimuthu Kalimuthu Daniel Musekamp Edgar Torres
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English
Semester:	SS 25
Organisation:	Maschinelles Lernen in den Simulationswissenschaften
Content:	https://www.f05.uni-stuttgart.de/informatik/studierende/seminaranmeldung/
Link:	C@MPUS

Intelligent Sensing and Perception

Learning with Less: Data- and Resource-efficient Visual Recognition

Advanced Seminar
2 Sh
WS 24/25

Number:	023800410
Lecturers:	Jun.-Prof. Dr.-Ing. Alina Roitberg
Type:	Advanced Seminar
Semester hours:	2
Teaching Language:	English/German
Semester:	WS 24/25
Organisation:	Autonomous Sensing and Perception Systems
Link:	C@MPUS

Foundation Models

Exercise
2 Sh
WS 23/24

Number:	023800405
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Jun.-Prof. Dr.-Ing. Alina Roitberg Univ.-Prof. Ph.D. Mathias Niepert
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	WS 23/24
Organisation:	Autonomous Sensing and Perception Systems
Content:	siehe Vorlesung
Link:	C@MPUS

Foundation Models

Lecture
2 Sh
WS 23/24

Number:	023800400
Lecturers:	Univ.-Prof. Dr. rer. nat. Steffen Staab Jun.-Prof. Dr.-Ing. Alina Roitberg Univ.-Prof. Ph.D. Mathias Niepert
Type:	Lecture
Semester hours:	2
Teaching Language:	German
Semester:	WS 23/24
Organisation:	Autonomous Sensing and Perception Systems
Content:	Foundation Models, such as GPT4 and Stable Diffusion, are models that are trained on very large amounts of data and can be effectively fine-tuned for a wide range of downstream tasks. This lecture aims to convey the key ideas behind the recent Large Language Models and Multimodal Foundation Models. The students will learn both, fundamental building blocks and optimization techniques, as well as advanced architectures and recent developments in the field of Foundation Models. Among others, the following topics will be covered: • Fundamentals: overview of the relevant building blocks and optimization techniques (e.g., transformer models, self-supervised learning), generative models. • Foundation models for text processing • Vision-based and multimodal foundation models • Discussion of ethical, societal, and legal facets of FMs Alongside the lectures, student projects are an integral part of this lecture. These projects focus on analysing and presenting recent research papers as well as implementing/using them for a specific use-case or application to gain hands-on experience. The students will present their results and write a report summarizing their learnings from the project.
Link:	C@MPUS

Foundation Models

Lecture
2 Sh
WS 24/25

Number:	023800400
Lecturers:	Jun.-Prof. Dr.-Ing. Alina Roitberg Univ.-Prof. Dr. rer. nat. Steffen Staab Univ.-Prof. Ph.D. Mathias Niepert Dr. rer. nat. Amin Totounferoush
Type:	Lecture
Semester hours:	2
Teaching Language:	German
Semester:	WS 24/25
Organisation:	Autonomous Sensing and Perception Systems
Content:	Foundation Models, such as GPT4 and Stable Diffusion, are models that are trained on very large amounts of data and can be effectively fine-tuned for a wide range of downstream tasks. This lecture aims to convey the key ideas behind the recent Large Language Models and Multimodal Foundation Models. The students will learn both, fundamental building blocks and optimization techniques, as well as advanced architectures and recent developments in the field of Foundation Models. Among others, the following topics will be covered: • Fundamentals: overview of the relevant building blocks and optimization techniques (e.g., transformer models, self-supervised learning), generative models. • Foundation models for text processing • Vision-based and multimodal foundation models • Discussion of ethical, societal, and legal facets of FMs Alongside the lectures, student projects are an integral part of this lecture. These projects focus on analysing and presenting recent research papers as well as implementing/using them for a specific use-case or application to gain hands-on experience. The students will present their results and write a report summarizing their learnings from the project.
Link:	C@MPUS

Foundation Models

Exercise
2 Sh
WS 24/25

Number:	023800405
Lecturers:	Jun.-Prof. Dr.-Ing. Alina Roitberg Univ.-Prof. Dr. rer. nat. Steffen Staab Univ.-Prof. Ph.D. Mathias Niepert Dr. rer. nat. Amin Totounferoush
Type:	Exercise
Semester hours:	2
Teaching Language:	English
Semester:	WS 24/25
Organisation:	Autonomous Sensing and Perception Systems
Content:	siehe Vorlesung
Link:	C@MPUS

Automotive and Assistive Computer Vision

Lecture
2 Sh
SS 24

Number:	023800000
Lecturers:	Jun.-Prof. Dr.-Ing. Alina Roitberg
Type:	Lecture
Semester hours:	2
Teaching Language:	English/German
Semester:	SS 24
Organisation:	Autonomous Sensing and Perception Systems
Content:	Computer vision is a vital technology in both automotive innovations like autonomous driving and in developing assistive technologies such as assistive robotics or tools for guiding the visually impaired. Despite rapid progress linked to the rise of deep learning, CV faces significant challeng-es in real-world applications, including handling adverse conditions, reli-ance on large annotated datasets, computational demands, understanding decision-making processes and uncertainty quantification. This module explores the fundamentals and advanced developments in CV, addressing these challenges with a focus on applications in dynamic environments. By examining both automotive and assistive technologies, students will learn about the synergies and shared obstacles between these fields. The curriculum covers essential CV techniques, efficient architectures, and domain adaptation, complemented by hands-on projects to apply what they've learned in practical settings. Among others, the following topics will be covered: • Fundamental methods and tasks: overview of the relevant build-ing blocks and optimization techniques (e.g., CNNs, visual trans-former models), tasks relevant for intelligent cars, e.g., obstacle detection and scene understanding, perception inside and out-side the vehicle; and assistive technologies (CV for improved mobility, activity recognition, document analysis) • Uncertainty in computer vision, open set recognition, visual recognition with limited training data • Domain adaptation/generalization and robustness against distri-butional shifts • Sensor fusion • Efficient architectures • Discussion of SoA architectures and synergies between automo-tive and assistive technologies, long-term implications. Alongside the lectures, student projects are an integral part of this lecture. These projects focus on analysing and presenting recent research papers as well as implementing/using them for a specific use-case or application to gain hands-on experience. The students will present their results and write a report summarizing their learnings from the project.
Link:	C@MPUS

Automotive and Assistive Computer Vision

Lecture
2 Sh
SS 25

Number:	023800000
Lecturers:	Jun.-Prof. Dr.-Ing. Alina Roitberg
Type:	Lecture
Semester hours:	2
Teaching Language:	English/German
Semester:	SS 25
Organisation:	Autonomous Sensing and Perception Systems
Content:	Computer vision is a vital technology in both automotive innovations like autonomous driving and in developing assistive technologies such as assistive robotics or tools for guiding the visually impaired. Despite rapid progress linked to the rise of deep learning, CV faces significant challeng-es in real-world applications, including handling adverse conditions, reli-ance on large annotated datasets, computational demands, understanding decision-making processes and uncertainty quantification. This module explores the fundamentals and advanced developments in CV, addressing these challenges with a focus on applications in dynamic environments. By examining both automotive and assistive technologies, students will learn about the synergies and shared obstacles between these fields. The curriculum covers essential CV techniques, efficient architectures, and domain adaptation, complemented by hands-on projects to apply what they've learned in practical settings. Among others, the following topics will be covered: • Fundamental methods and tasks: overview of the relevant build-ing blocks and optimization techniques (e.g., CNNs, visual trans-former models), tasks relevant for intelligent cars, e.g., obstacle detection and scene understanding, perception inside and out-side the vehicle; and assistive technologies (CV for improved mobility, activity recognition, document analysis) • Uncertainty in computer vision, open set recognition, visual recognition with limited training data • Domain adaptation/generalization and robustness against distri-butional shifts • Sensor fusion • Efficient architectures • Discussion of SoA architectures and synergies between automo-tive and assistive technologies, long-term implications. Alongside the lectures, student projects are an integral part of this lecture. These projects focus on analysing and presenting recent research papers as well as implementing/using them for a specific use-case or application to gain hands-on experience. The students will present their results and write a report summarizing their learnings from the project.
Link:	C@MPUS

Automotive and Assistive Computer Vision

Exercise
2 Sh
SS 24

Number:	023800005
Lecturers:	Jun.-Prof. Dr.-Ing. Alina Roitberg
Type:	Exercise
Semester hours:	2
Teaching Language:	English/German
Semester:	SS 24
Organisation:	Autonomous Sensing and Perception Systems
Link:	C@MPUS

Automotive and Assistive Computer Vision

Exercise
2 Sh
SS 25

Teaching Overview

Analytic Computing

Probabilistic Machine Learning

Probabilistic Machine Learning

Oberseminar Artificial Intelligence

Machine Learning with Graphs

Laboratory Course Artificial Intelligence: Deep Learning Lab

Foundation Models

Foundation Models

Formal Methods for the Web

Socially Intelligent Robotics Lab (Autonomous Systems)

Oberseminar Intelligent Robotics

Mobile Robotics

Mobile Robotics

Advanced seminar “Research Topics for Intelligent Interactive Robots”

Machine Learning for Simulation Science

Theoretical and Methodological Foundations of Autonomous Systems

Theoretical and Methodological Foundations of Autonomous Systems

Theoretical and Methodological Foundations of Autonomous Systems

Theoretical and Methodological Foundations of Autonomous Systems

Reinforcement Learning

Reinforcement Learning

Reinforcement Learning

Reinforcement Learning

Recent Applications of Machine Learning

Recent Applications of Machine Learning

Machine Learning for Simulation Science

Machine Learning for Simulation Science

Machine Learning for Simulation Science

Kolloquium Machine Learning for Simulation Science

Kolloquium Machine Learning for Simulation Science

Grundlagen der Künstlichen Intelligenz

Grundlagen der Künstlichen Intelligenz

Grundlagen der Künstlichen Intelligenz

Grundlagen der Künstlichen Intelligenz

Foundations of Artificial Intelligence / Theoretical and Methodological Foundations of Autonomous Systems

Foundations of Artificial Intelligence / Theoretical and Methodological Foundations of Autonomous Systems

Deep Learning for the Sciences

Deep Learning for the Sciences

Deep Learning for the Sciences

Deep Learning for the Sciences

Deep Learning for the Sciences

Intelligent Sensing and Perception

Learning with Less: Data- and Resource-efficient Visual Recognition

Foundation Models

Foundation Models

Foundation Models

Foundation Models

Automotive and Assistive Computer Vision

Automotive and Assistive Computer Vision

Automotive and Assistive Computer Vision

Automotive and Assistive Computer Vision

Here you can reach us

Audience

Formalities

Services

Organization