Causal inference is essential for achieving responsible, trustworthy AI and advancing public health. Two foundational tasks form its basis: determining whether causal effects can be computed from available data (causal effect identification) and then actually computing these effects (causal effect estimation). While the theory of causal effect identification is well-established, effective and robust estimation remains largely unsolved, and addressed for only a subset of identification scenarios. Fully connecting identification with estimation makes causal inference far more practical, ultimately driving trustworthy AI and improving public health through causal reasoning.

In this talk, I will present my work that unifies estimation strategies across diverse practical scenarios. I will begin by introducing an estimation framework for any identifiable causal effect from observational data, demonstrating its application in public health and explainable AI. Building on this foundation, I will describe an extended estimation framework that leverages a fusion of observational and experimental datasets, thereby enabling robust estimation in more complex settings. I will further present a unified causal effect estimation method that encompasses a range of causal inference scenarios, including applications in fairness and decision making. Finally, I will conclude with a brief discussion of future research directions aimed at advancing healthcare science and trustworthy AI through innovations in causal inference.

Yonghan Jung is a PhD candidate in Computer Science at Purdue University and a member of the CausalAI Lab at Columbia University, led by Professor Elias Bareinboim. He works in causal inference, developing robust causal effect estimators essential for responsible and trustworthy AI and applicable to public health. His research has been published at top AI venues such as NeurIPS, ICML, and AAAI. He has demonstrated a causal inference pipeline for public health at the American Thoracic Society (ATS) and co-hosted a tutorial at FAccT 2022 on understanding deep neural networks through the lens of causality. During his PhD, he interned as an applied scientist at Amazon’s Causality team, where he developed causality-based explainable AI methods. More information can be found at https://www.yonghanjung.me/