Innovation Project in Conversational AI (Master) (DS-IPro-M)

Conversational AI combines methods for Natural Language Processing and Machine Learning to communicate with users. To aquire theoretical knowledge, participants will examine current work and system prototypes of the group. Subsequently, they will gain practical experience by implementing extensions to these prototypes individually or in small groups, depending on the scale of the selected topic. At the end of the semester, participants will demonstrate their implementations in oral presentations and submit term papers.

Conversational AI includes a broad range of topics, such as:

• Dialogue Systems
• Deep Learning
• Large Language Models
• Retrieval-Augmented Generation
• Knowledge-Learning

The projects deal with the (further) development in one of aforementioned topics in innovative use-case scenarios based on current prototypes of the Natural Language Generation and Dialogue Systems Group. The projects are coordinated individually with students.

The learning goals of this course are the following: Students will learn to familiarize themselves with a complex project topic in the field of Conversational AI. They will be able to investigate related work and then apply and implement their topic in practice. The students will learn to appropriately present their work in the format of a scientific paper and oral presentation.

The project offers the opportunity to work on an individual topic in the field of innovative Conversational AI. For the implementation and project work, students are expected to use Python and git. Further tools and libraries will be discussed with a supervisor. The number of participants is limited. Typical Workload:

• Meetings and talks: ~25h
• Familiarization with the project: ~30h
• Implementation: ~120h
• Preparation of presentation and report: ~35h

Please send your application for the seminar before the start of the semester to Nicholas Walker or Nicolas Wagner.