Large Language Models (LLMs) for IT Practitioners, 3 Credits

Contents

This course is intended for IT practitioners that would like an organized introduction to the concepts and tools underlying Large Language Models (LLMs). We cover Transformers, BERT and GPT at a high level and then explore few-shot approaches, RAG approaches using LangChain and Vector Databases and finally approaches to using LLMs to support dialogue over SQL databases. Some parts of the examination will let students develop solutions over their own industry-specific data.

Expected learning outcomes

Knowledge and understanding
After completing the course, the student should be able to:

• (FSR 1) Explain the components, processes and workflow of building GPT-based models.
• (FSR 2) Describe prompt engineering, few-shot, vector database and RAG-based approaches to using LLMs.

Competence and skills
After completing the course, the student should be able to:

• (FSR 3) Set up and run simple GPT learning problems using PyTorch.
• (FSR 4) Use LangChain and vector databases to implement RAG and advanced RAG pipelines.
• (FSR 5) Set up few-shot approaches to building dialogue systems over SQL databases.

Judgment and approach
After completing the course, the student should be able to:

• (FSR 6) Understand and discuss concepts and terminology of state-of-the-art LLMs.
• (FSR 7) Develop an ability to distinguish fact from fantasy in this fast moving field.

Required Knowledge

The eligibility requirements include 6 ECTS introduction to programming and 6 ECTS data structures and algorithms as well as 30 ECTS in Mathematics or Computer Science. Proficiency in Python programming is strongly recommended.

Form of instruction

This course is a distance course which requires no local physical presence. Lectures are held over Zoom and course material is delivered via Canvas. Lectures are recorded, so attendance is flexible. Students answer problems and demonstrate their programming solutions in recitation sessions held over Zoom. While attendance in recitations is mandatory, there are multiple opportunities to accommodate student schedules. The course ends with a take home exam that students upload to Canvas. All work is conducted by the student individually.

Examination modes

The examination consists of three Zoom-based recitations and a final take-home exam. Some parts of the examination will let students develop solutions over their own industry-specific data.

The grade scale is Fail (U), Pass (G), and Pass with distinction (VG). A passing grade requires passing all parts of the examination. The grade is set after a combined assessment of all parts of the examination.

Adapted examination
The examiner can decide to deviate from the specified forms of examination. Individual adaptation of the examination shall be considered based on the needs of the student. The examination is adapted within the constraints of the expected learning outcomes. A student that needs adapted examination shall no later than 10 days before the examination request adaptation from the Department of Computing Science. The examiner makes a decision of adapted examination and the student is notified.

Other regulations

If the syllabus has expired or the course has been discontinued, a student who at some point registered for the course is guaranteed at least three examinations (including the regular examination) according to this syllabus for a maximum period of two years from the syllabus expiring or the course being discontinued.