Trainers:
Mafalda Dias and Jonathan Frazer are group leaders at the CRG, Barcelona, where they co-lead the Probabilistic Machine Learning and Genomics group.
They develop deep learning and probabilistic modeling approaches applied to genomics and other large-scale multimodal biological data to address questions in disease genetics, protein function and conservation genomics.
Course objective:
This course aims to develop a principled understanding of probabilistic model building in the context of modern AI, equipping participants to reason under uncertainty, integrate prior knowledge, and construct interpretable, data-efficient models. It encourages a shift from purely data-driven approaches to more structured and hypothesis-aware modelling, supporting better decision-making and experimental design.
Learning outcomes:
- Apply probabilistic reasoning to model building in AI
- Incorporate prior knowledge into structured, interpretable models
- Understand key probabilistic approaches in modern machine learning
- Design more efficient experiments using active learning principles
- Critically assess modelling assumptions and limitations
What to bring:
- A laptop
- Active Gmail account (we will use Google Collab environments)
