Machine learning and causal inference

Introduction to machine learning (from a computational biology perspective)

Note

Aims:

• get intuition and high-level understanding of what machine learning is and what types of problems it can help solving;

• this material will also show that machine learning is not just a black box, but that different choices have important implications;

• provide the overall understanding that there are a data representation component and a machine learning algorithm;

• achieve high-level understanding of machine learning workflow, comparison and the related uncertainty

Level: beginner 🌱

If you are new to machine learning and want to apply it in a biological discipline, the following material might provide a convenient introduction. It consists of 6 parts and combines the content and references with tasks and quizzes to test the covered material.

Topics:

• Brief overview of machine learning in computational biology
• Introduction to Machine Learning
• Introduction to machine learning models and algorithms
• Data representation in machine learning
• Machine learning model comparison and uncertainty
• Transparency and reproducibility in machine learning

Recommended material

Recommended courses at UiO:

• IN5400/IN9400 Machine Learning for Image Analysis 🌿

• STK-IN9300 Statistical Learning Methods in Data Science 🌿

Recommended seminars at UiO:

• Research Seminar at the Section for Machine Learning, Department of Informatics

Recommended books:

• Deep Learning book by Goodfellow et al. 2016

• The Elements of Statistical Learning by Hastie et al. 2008

Relevant papers:

• Model Evaluation, Model Selection, and Algorithm Selection in Machine Learning by Raschka 2018

• DOME: recommendations for supervised machine learning validattion in biology by Walsh et al. 2021

• An introduction to domain adaptation and transfer learning by Kouw and Loog 2018

• Causality for Machine Learning by Schölkopf 2019

• Causal Inference and the Data-Fusion Problem by Bareinboim and Pearl 2016

• Towards Causal Representation Learning by Schölkopf et al. 2021