Causality and Machine Learning

Catalog description

This course aims to bridge the gap between two powerful fields, machine learning and causal inference. While machine learning has achieved significant success in various domains, the lack of trustworthiness (e.g., explanation, fairness, generalization, and robustness) hinders its widespread adoption. Causal inference, on the other hand, is a vital discipline that explores cause-and-effect relationships, going beyond simple correlations within data systems. By understanding causal relationships, we can uncover the essence of artificial intelligence, fostering trustworthy machine learning practices. In this course, we will provide a foundation in both traditional causal inference and machine learning concepts. Moreover, we will explore the recent advancements in combining these two areas and highlight the mutual benefits they gain from each other. The course covers mathematical skills (e.g., graphical models and probabilistic reasoning) and cutting-edge machine learning & neural network techniques. Offered as CSDS 352 and CSDS 452 . Prereq: CSDS 340 or CSDS 440 .