Course Slides
The following material consists of the course slides used for the Artificial Intelligence and Society course. Each module is accompanied by the respective Python tutorial (notebook), that you can find in Python Tutorials.
-
Responsible AI
Learn more about the pillars of Responsible AI development and how to operationalize Trustworthy AI.
-
Data-Centric AI
Explore the differences between Model-Centric AI and Data-Centric AI and the main data imperfections that can arise in real-world domains.
-
Data Complexity
How complex is you classification problem? Learn about different sources of data complexity and how to measure them effectively.
-
Imbalanced Data
Real-world domains are often imbalanced. Discovery the main strategies to mitigate class imbalance.
-
Bias and Fairness
There are many ways in which bias can creep into our AI pipelines. Learn how to identify, measure, and mitigate biased predictions.
-
Missing Data
Missing data arises in several domains, and presents different mechanisms. Learn how to overcome this problem in practical applications.
-
Explainability
Responsible AI requires models to be understandable and auditable. Explore different strategies to generate and interpret explanations for model predictions.
-
Data Privacy and Synthetic Data
How can we learn about trends and behaviours in sensitive data, while protecting information that is specific to individuals? Can synthetic data be a solution?