Causal AI

Explore causal AI to understand the differences between correlation and causation in machine learning. This lesson explains how causal models, structural equation models, and Bayesian networks establish cause-effect relationships while addressing confounding variables through methods like propensity score matching. Learn the advantages and complexities of causal AI for more explainable and robust ML applications.

We'll cover the following...

Causal models
Advantages of causality
Disadvantages of causality

One of the problematic aspects of ML is that most of it is completely correlational, not causal. ML algorithms work primarily by applying statistical inference and reasoning using a series of variables and a target. In essence, it tries to identify whether variable $A$ occurs with target $B$ . If so, it may be that “ $A$ predicts $B$ .” However, there are many reasons why $A$ is predictive of $B$ . One particularly dangerous outcome is when there’s a lurking variable present in the data. A lurking variable is an unreported variable $(C)$ that makes it seem like ...

1.Introduction

2.Disasters in Data

3.Disasters in Models

Project

4.Alternatives to Traditional ML

Project

5.Conclusion

Assessment

Causal AI