Causal Artificial Intelligence

CHAPTER 4Creating Practical Causal AI Models and Systems

In this chapter, we will link DAGs to SCMs and how they work in tandem to create a practical and functional causal AI model. Therefore, we will discuss how to turn a causal model based on DAGs and SCMs into a practical and functioning causal AI system. In Chapter 3, we compared the two families of AI: the newly emerging causal-based AI and the more traditional correlation-based AI. We also introduced the foundational concepts of understanding causal AI model building. We talked about basic directed acyclic diagrams (DAGs), and we built a simple model to illustrate the introductory concepts that we are interested in and will be using and building upon throughout the remainder of the book. In this chapter, we will focus on detailing and expanding upon the formal language and methodology used in describing, designing, building, modifying, validating, and leveraging causal models.

Understanding Complex Models

There have been, and are, numerous approaches to naming the elements of causal models, discussing modeling, and communicating causal concepts. One of the impressive aspects of causal models is that the models and elements of causal models can be used to visually describe any model including variables, relationships, processes, and environments that are important for a hybrid team. Any environment, process, or interaction of variables can be described and modeled in a causal-based approach.

If you want to understand ...