Type of course

The course can be taken as a singular course.

Registration is open for UiT students, members of NORA Research School and NLDL participants.

It will be conducted as a concentrated course in the style of winter school 5th to 9th Jan 2026.

Admission requirements

PhD students or holders of a Norwegian master´s degree of five years (300 ECTS) or 3 (180 ECTS) + 2 years (120 ECTS) or equivalent may be admitted.PhD students must upload a document from their university stating that there are registered PhD students. This group of applicants does not have to prove English proficiency and are exempt from semester fee. Holders of a Master´s degree must upload a Master´s Diploma with Diploma Supplement /English PhD students at UiT The Arctic University of Norway register for the course through StudentWeb. External applicants apply for admission through SøknadsWeb. All external applicants have to attach a confirmation of their status as a PhD student from their home institution. Students who hold a Master of Science degree, but are not yet enrolled as a PhD-student have to attach a copy of their master’s degree diploma. These students are also required to pay the semester fee.

All students must register for the Northern Lights Deep Learning Conference, further information about this process can be found here.

Recommended prerequisites: Programming skills in python and hands on knowledge of python programming for deep learning. Knowledge of machine learning at master’s level from study programs in computer science, physics and technology, mathematics and statistics, or equivalent.

Application code: 9316, application from the1st of October to deadline the 1st of November.

The course is limited to 40 participants, with a first come, first serve admission for qualified applicants. If there are more than 40 applicants, a waiting list will be created and the course resonsible will evaluate if the capacity should be increased.

Course content

This course will provide a study of several emerging topics of high relevance within advanced deep learning, from a basic understanding of the techniques to the latest state-of-the-art developments in the field.

The course will consist of 5 full days of the NLDL conference including: tutorials, keynote sessions, and practical components.

Core concepts:

• Large language models
  • Copilots
  • AI agents
• Non-Euclidean deep learning
  • Hyperbolic learning
• Multimodal learning
  • Multimodal information fusion
  • Text-vision modeling
  • Multimodal loss functions
• Ethical AI
  • Ethical challenges with AI in natural language processing.
  • Bias in AI.
  • Fairness in AI

Relevance of course in program of study: Deep learning has revolutionized technology and science by enabling information to be extracted with much better precision and at much larger scales compared to only a few years ago from data sources such as images, text, speech, biological material, chemical components, and sensory measurements in general. Most study programs offer courses on the fundamental theory and applications of neural networks and machine learning systems in general, which provide the backbone of deep learning. The course covers advanced topics in artificial intelligence, including large language models (LLMs), multimodal learning, non-Euclidean deep learning, and ethical AI. It provides an in-depth understanding of LLMs, focusing on their use with copilots and AI agents. Multimodal learning is explored through methods that integrate information from diverse data sources such as text, images, and audio, highlighting its impact on real-world tasks. The curriculum also delves into non-Euclidean deep learning, examining techniques for processing data with non-Euclidean geometry like tree structures. Ethical AI is addressed through discussions on fairness, transparency, and accountability, with practical approaches for embedding ethical principles into intelligent systems. Throughout the course, students engage with state-of-the-art research and hands-on exercises to apply these concepts in practical scenarios.

Objectives of the course

• Knowledge - the student is able to
  • describe large language models (LLMs) and their foundational concepts.
  • discuss pretraining, fine-tuning, and practical applications of LLMs.
  • explore multimodal learning and integration of diverse data types (e.g., text, images, audio).
  • discuss architectures and techniques for multimodal fusion in machine learning.
  • describe non-Euclidean deep learning and its relevance to hyperbolic learning.
  • discuss methods and challenges in processing non-Euclidean structures with deep learning.
  • describe ethical AI principles and frameworks.
  • discuss fairness, transparency, accountability, and approaches for embedding ethics in AI systems.
• Skills - the student is able to.
  • explain methods for processing and analyzing non-Euclidean structures in deep learning.
  • explain ethical AI principles and frameworks.
  • explain fairness, transparency, accountability, and approaches for embedding ethics in AI systems.
  • apply large language models, multimodal learning, non-Euclidean deep learning, and ethical AI concepts to new application or problem settings.
  • use multimodal, non-Euclidean, and ethical AI techniques in research and industrial settings with software libraries such as PyTorch.
• General competence - the student is able to
  • give an interpretation of recent developments and provide an intuition of the open questions in the field of natural langauge processing, multimodal learning, and non-Euclidean learning.
  • show an understanding of Ethical AI (RAI), and how to understand ethical, legal, and societal impacts, and methods to embed ethical principles into AI systems.

Language of instruction and examination

Teaching methods

Lectures: 40 hours (full 5 days of the winter school)

Self-study sessions: 40 hours

Project work: spread over 8 weeks - net time 50 hours

Hands-on sessions: 10 hours

Net effort (~120 hours)

Final exam