Administrative Information
| Title | Decision networks |
| Duration | 60 |
| Module | A |
| Lesson Type | Tutorial |
| Focus | Technical - Foundations of AI |
| Topic | Decision networks |
Keywords
Naive Bayes-net,Bayes-net,Decision network,optimal decision,value of information,
Learning Goals
- Student can design an interventional experiment for A/B testing and an imagery evaluation of A/B testing
- Student can construct causal diagrams and general decision nets
- Student can test fairness using a causal model
Expected Preparation
Learning Events to be Completed Before
Obligatory for Students
- Artificial Intelligence: A Modern Approach, 4th Global ed. by Stuart Russell and Peter Norvig, Pearson (AIMA4e):ch16-17
- concepts of probability
- axioms of probability theory
- concept of independence
- Bayes' rule
- Bayesian model averaging
- universal AI
- multivariate joint probability distributions
- chain rule
Optional for Students
- AIMA4e:ch16-17
References and background for students
- AIMA4e:ch16-17
Recommended for Teachers
- AIMA4e:ch16-17
- Charniak, E., 1991. Bayesian networks without tears. AI magazine, 12(4), pp.50-50.
- Pearl, J., 2019. The seven tools of causal inference, with reflections on machine learning. Communications of the ACM, 62(3), pp.54-60.
Lesson materials
Instructions for Teachers
In parallel with the lecture build a loan evaluation decision net model and investigate its fairness.
Outline/time schedule
| Duration | Description |
|---|---|
| 15 | Explain a causal diagram |
| 15 | Set evidences and discuss optimal action. |
| 15 | Identify a variable E' with maximal value of information. |
| 15 | Draw a random value from E' and select the optimal action. |
Acknowledgements
The Human-Centered AI Masters programme was Co-Financed by the Connecting Europe Facility of the European Union Under Grant №CEF-TC-2020-1 Digital Skills 2020-EU-IA-0068.