In " Hot Research Topics in AI for Engineering Applications", we explore the applicability of cutting-edge research findings in the fields of Machine Learning and Artificial Intelligence (e.g., LLM agents, Reinforcement Learning) to applications in engineering (e.g., optimization in production and logistics, creation of CAD models). Each year, we offer a different methodological focus (more on the IMI-homepage).

First, we provide the theoretical foundations and then move into a group work phase where students implement and analyze an application prototype. The event is aimed at students with prior knowledge in machine learning and programming.

The topics of this course:

• Theoretical foundations of the technologies considered in the course (e.g., Deep Learning, Transformers, LLM)
• Application possibilities of modern technologies in an industrial context
• Challenges in making current research findings usable for solving specific engineering problems and productive use
• Implementation of solutions to apply modern technologies to specified engineering problems (usually Python-based, using current frameworks)
• Independent execution of an implementation project with current, thematically relevant content (e.g., LLM agents for interaction with external systems such as robots, for algorithm construction, or for creating 3D CAD models, etc.)
• Technologies and applications are announced at the beginning of each semester

After the event, participants will be able to:

• Identify the technical and algorithmic foundations behind the relevant research topics and explain their functionalities
• Identify application possibilities of current research findings and related technologies in an industrial context, as well as the challenges that arise in the process
• Implement solutions proposed in recent publications using existing frameworks and codebases as prototypes
• Structure and execute programming projects in a team
• Clearly present the results of practical projects tailored to the audience

Participation Requirements

A foundational course in Machine Learning, Data Science, or Deep Learning, ideally including practical application in Python, must have been successfully completed.

Recommended courses include:

• T-MACH-114078 – Data Science Fundamentals in Engineering
• T-ETIT-109839 – Labor für angewandte Machine Learning Algorithmen
• M-INFO-106014 – Grundlagen der Künstlichen Intelligenz
• T-WIWI-110340 – Angewandte Informatik – Anwendungen der Künstlichen Intelligenz

Comparable completed courses may be accepted as proof of the required foundational knowledge.

Additional Information

Due to capacity limitations, the number of participants is limited to 20. If necessary, a selection process will be conducted. Places will be allocated based on the students’ academic progress. Details on this, as well as information on how to apply, will be posted on ILIAS.

Organizational Information

See further details on course organization at: teaching.imi.kit.edu and in ILIAS.

Workload

120 hours

Place and time of the course can be found in ILIAS, / Ort und Zeit der Lehrveranstaltung siehe ILIAS