Advanced Machine Learning and AI for Life

Course goals

Intended Learning Outcomes (ILO):
At the end of the course, you should be able to:
a) Develop a machine learning model by fine-tuning a foundation model
b) Analyse and evaluate AI-based predictions using explainable AI methodology
c) Understand and evaluate how multi-modal models make predictions and integrate different data types
d) Design a machine learning task, given data and a research question, within the field of the Life Sciences

Content

DESCRIPTION OF CONTENT
This master’s-level course explores cutting-edge applications of artificial intelligence in the life sciences, with a focus on foundation models, explainable AI (XAI), and multi-modal machine learning. Students will gain a deep understanding of how large-scale foundation models, we will work with protein language models, can be fine-tuned to prediction aggregation properties of proteins. In addition, you will learn to make predictions about LC-MS/MS experiments, using explainable AI. Lastly, we will cover multi-modal prediction models through journal club-style discussions.
The course will be a mixture of 8 lectures, 6 practical classes and 2 journal club sessions, followed by an individual (oral) assessment.

TARGET GROUP
The course is aimed at students either:
- with a Life Science background, who have already participated in a basic machine learning course, and would like to learn about more complex AI and Machine Learning techniques.
- or for students with a Computational Science background who have already participated in a basic machine learning course and would like to gain a deeper understanding of complex AI problems in the Life Sciences

Mandatory for students in Master’s programme: No.
Optional for students in other Master’s programmes GS-LS AND Computer and Information Sciences, AI and Data Science: YES

ASSESSMENT
The course contains the following assessment elements:
Assignment 1 – Explainable AI  (ILO: b, d); asssement type: pass/fail and formative feedback.
Assignment 2 - Foundation model finetuning (ILO: a, d); asssement type: pass/fail and formative feedback.
Journal Club (Multi-modal AI, ILO b, d); asssement type: pass/fail and formative feedback.
Final oral exam (ILO: a, b, c, d) — scheduled on the course last day; assessment type: rubrics based on ILOs; 100% weight of the final grade;