| Code |
16686
|
| Year |
3
|
| Semester |
S1
|
| ECTS Credits |
6
|
| Workload |
PL(30H)/T(30H)
|
| Scientific area |
Informatics
|
|
Entry requirements |
N/A
|
|
Learning outcomes |
The course has two main objectives. On the one hand, it aims to provide an introduction to the field of Interpretability in Artificial Intelligence by presenting the main techniques for explaining the decision-making process of this type of machine learning methods. On the other hand, it aims to introduce the main concepts of Causality, useful in the field of Data Science to determine the causes of a given result, based on the design of causal inference protocols from observational data.
At the end of the course, students should be able to:
a. Understand the main interpretability methods that allow generic or specific machine learning models to be explained;
b. Understand the fundamental principles of causality that allow causal inference protocols to be designed
c. Understand the main methods of causal inference based on observational data.
|
|
Syllabus |
A. Introduction to Explainable AI (XAI): Motivation, fundamental concepts, and categorization of different approaches.
B. Global techniques for explaining the decision-making process of machine learning models: Ceteris Paribus, ICE, PDP, ALE.
C. Local techniques for explaining the decision-making process of machine learning models: LIME; Shapley Values; SHAP.
D. Techniques for explaining specific models such as convolutional neural networks.
E. Introduction to Causality and history of different causal models (Potential Outcomes Framework and Graph-based Causal Models)
F. Causal Inference: Main strategies for identifying causal relationships in sets of variables: RCTs, backdoor adjustment, instrumental variables, DiD, discontinuous regression.
G. Causal graphs and structural causal models.
H. Applications of Causality in Machine Learning Problems
|
|
Main Bibliography |
Molnar, C. (2025). Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (3rd ed.).
Hernán MA, Robins JM (2020). Causal Inference: What If. Boca Raton: Chapman & Hall/CRC
Facure, M. (2021). Causal Inference for The Brave and True. Retrieved from https://matheusfacure.github.io/python-causality-handbook/
Pearl, J. (2009). Causality: Models, Reasoning, and Inference. Cambridge University Press
Pearl, J., & Mackenzie, D. (2018). The Book of Why: The New Science of Cause and Effect
|
|
Language |
Portuguese. Tutorial support is available in English.
|