UGA MSIAM Learning, Probabilities and Causality
Teachers:
Charles K. Assaad, Emilie Devijver, Eric Gaussier
Description:
Causality is at the core of our vision of the world and of the way we reason. It has long been recognized as an important concept and was already mentioned in the ancient Hindu scriptures: “Cause is the effect concealed, effect is the cause revealed”. Even Democritus famously proclaimed that he would rather discover a causal relation than be the king of presumably the wealthiest empire of his time. Nowadays, causality is seen as an ideal way to explain observed phenomena and to provide tools to reason on possible outcomes of interventions and what-if experiments, which are central to counterfactual reasoning, as ''what if this patient had been given this particular treatment?’’. In this lecture, we will provide an overview of causality, from its first definitions centuries ago to its modern usage in machine learning and reasoning. In particular, we will answer the following questions:
- - How to represent causal relations through structural causal graphs?
- - How to infer causal relations from purely observational data, from purely interventional data and from a mixture of them?
- - How to exploit and reason upon causal knowledge? In particular, can one quantify the relation between a cause and its effect? Can one compute the effect of an intervention? Can one use causal knowledge for counterfactual reasoning or mediation analysis?
Remark: The course will be divided in lectures and pratical sessions aimed at better understanding the different notions introduced. The concepts behind causality are not too diificult to grasp but nevertheless differ from traditional probability concepts.
*What-if experiments are central to counterfactual reasoning.
Prerequisite:
Probability and statistics background.
Selected books (the important ones are in italics):
- The Book of Why: The New Science of Cause and Effect, by Pearl and Mackenzie, 2018
- Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, by Pearl, 1988
- Causation, Prediction, and Search, by Spirtes, Glamour and Scheines, 2000
- Elements of Causal Inference: Foundations and Learning Algorithms, by Peters, Janzing and Scholkopf, 2017
- Causality: Models, Reasoning and Inference, by Pearl, 2009
Materials:
| Date | Description | Slides |
|---|---|---|
| November 25 | Introduction to causal graphical models | Download |
| December 2 | Structural equation models, structural causal models | |
| December 2 | Causal discovery: constraint-based methods | Download |
| December 9 | Causal discovery: score-based and noise-based methods | Download |
| January 4 | Lab on Graphs | Download |
| January 4 | Graded Lab on PC | Download |
| January 6 | Back-door and front-door criterions | Download | January 6 | Do-calculus and identifiability | January 13 | Causality and machine learning | January 13 | Graded Lab on Simpson’s paradox |
| TBD | Lab on introduction to graphs and PC algorithm |