Tools, Practices, and Techniques to Enable Causal AI

Summary

This chapter is dedicated to providing a guide to the tools, technologies, and best practices for implementing causal artificial intelligence (AI) solutions. It proposes a causal AI pipeline, which aims to identify steps along the way to implementing causal AI within our organization. From data identification and model design and development to deployment and continuous learning, the chapter delves into existing software available for each step of the pipeline. In addition to the causal AI pipeline and tools, it discusses some specific techniques being used in causal AI, specifically the creation of synthetic data. Model development is the core of the causal AI pipeline, where data scientists, domain experts, and hybrid team work together to design and implement the causal AI model. This stage involves selecting appropriate causal inference techniques using visual models.