The achievements of predictive machine learning have made clearer than ever that black-box solutions cannot satisfactorily inform human decision-making in critical areas such as medicine, economics and government policy. Because of this, we are slowly but steadily observing a shift, both in academia and industry, towards techniques that offer transparency and explainability. Causal Machine Learning (ML) represents the field of research in which algorithms aim to recover some form of causal representation or model from data. In this talk, I will explain why causal ML is emerging as a crucial approach to complement predictive ML, and share my thoughts on why adoption of these approaches has been slow, with little to moderate impact in practice.
Zoom link: https://zoom.us/j/97374751400?pwd=UjZ4cUVaSXZ2bnZUT1BNSVhySFlDQT09
