One of the biggest problems faced by machine learning today is that so much of real-world data is not generated in a way which we usually use to train the Artificial Intelligence (AI) models, and this can be addressed by the use of causal inference in machine learning.
When we try rationalising various things, we think in terms of how something causes the other. If we understand why certain things happen, we can put in efforts to improve the future results/outcomes. Causal inference is a statistical tool that enables the AI and machine learning algorithms to reason similarly. Randomised Controlled Trials (RCT) are the gold standard for inferring the causal effects. Causal inference enables us to understand how some variables affect others. In this particular session, we explored the idea of using machine learning methods for causal inference from observational data / real world evidence data.