Title: Storytelling with Data in Causal Machine Learning
Short Code: TA.MSE_SDCM
ECTS Credits: 3
UAS: HSLU
Organizer Details: HSLU, Institute Natural Sciences and Humanities
Evaluation:

Written project summary and oral group presentation.

 

Decision Date: 1 September 2024 
Start Date: 9 September 2024 
End Date: 13 September 2024 
Date Details:
Type:
Language(s):

EN

Description (max. 300 characters):

In today's data-driven world, the ability to effectively communicate insights and findings from data analysis is a critical skill. This course aims to equip students with the necessary knowledge and skills to effectively communicate complex causal relationships through data storytelling and data visualisation.

The first part of the block course will cover the fundamental concepts of causal inference, including randomised controlled trials, observational studies, counterfactuals, confounding, and causal graphs. Students will learn the principles and techniques of various causal machine learning topics, including causal supervised learning, causal explanations, and causal fairness. The second part of the block course will focus on effective data visualisation and storytelling techniques to communicate complex causal relationships. 

Throughout the course, participants will engage in hands-on exercises and projects that will allow them to apply the concepts and techniques learned to analyze, interpret, and explain causal relations in real-world datasets. By the end of the course, participants will have the skills and knowledge necessary to analyse and communicate causal relationships through data storytelling and visualisation, a valuable skill in a range of fields, including data science, medicine, engineering and economics.

Contents and Learning Objectives:


 

Monday

Sep 9 2024

Tuesday

Sep 10 2024

Wednesday

Sep 11 2024

Thursday

Sep 12 2024

Friday

Sep 13 2024

8:30-10:00

Overview of causality concepts in Physics, Biology, Medicine, and Economics

Recap of Concepts of Causality, Discussion

Recap of causal supervised learning, explanations, and fairness, Discussion

Visualisation techniques in Statistics

Recap of visualisation techniques and data storytelling, Discussion

10:00-10:30

Break

Break

Break

Break

Break

10:30-12:00

Self-contained introduction into causal inference

Causal Supervised Learning

Presentation of topics for course project and of available datasets from Physics, Medicine, Economics

Visualisation Techniques in Machine Learning

Independent work on course project

12:00-13:00

Break

Break

Break

Break

Break

13:00-14:30

Causal Bayesian Networks, Structural Causal Models

Causal Explanations

Independent work on course project

Data Storytelling

Independent work on course project

14:30-15:00

Break

Break

Break

Break

Break

15:00-16:30

Causal Representation Learning, Spurious Relationships due to Confounding, Causal Influence

Causal Fairness

Independent work on course project

Data Storytelling

Short group project presentation with feedback for course project

17:00-18:00

Backup

Backup

Short group project presentation with feedback for course project

Backup

Recapitulation, feedback, evaluation, and closing





































Professional competencies: The students…

F1: are able to explain the importance of causality in physics, biology, medicine, and economics.

F2: are able to explain the main concepts of causal inference such as interventions and counterfactuals and to formalize causal relations.

F3: are able to apply appropriate causal machine learning methods such as causal supervised learning, causal explanations, and causal fairness to a given problem class.

F4: are able to apply powerful visualization techniques to highlight causal relations for a given dataset and a problem setting.

F5: are able to apply powerful storytelling techniques to highlight causal relations for a given dataset and a problem setting.

Methodological competencies: The students…

M1: are able to choose and apply causal machine learning tools and methods correctly for a given dataset and problem class.

M2: are able to critically question the results and draw the correct conclusions from causal machine learning tools and methods.

M3: are able to explore, present and visualize causal relations in datasets with clarity and conviction.


Admission:
Literature:

Lecture Notes, Slides, and Code Repository will be available.

Conditions:
Contact:

Prof. Dr. Mirko Birbaumer

 
Contact Person E-Mail: mirko.birbaumer@hslu.ch
Status: registration open
 
Specialization: Business Engineering and Production (BEP)

Energy and Environment (EE)

Industrial Technologies (InT)

Information and Communication Technologies (ICT)

Geomatics (GEO)

Aviation (Avi)

Business Engineering (BE)

Civil Engineering (CE)

Computer Science (CS)

Data Science (DS)

Electrical Engineering (ElE)

Energy & Environment (EnEn)

Geomatics (Geo)

Mechanical Engineering (ME)

Mechatronics & Automation (MA)

Medical Engineering (Med)

Photonics (Pho)

 

[Responsible for this text: Birbaumer Mirko]