Supervised Causal Discovery and Its Applications in Data Management

PhD Qualifying Examination


Title: "Supervised Causal Discovery and Its Applications in Data 
Management"

by

Mr. Pingchuan MA


Abstract:

Understanding causal relations is one of the most fundamental problems in 
scientific discovery, such as clinical trials, economics. The gold 
standard for inferring causal relations is to conduct randomized 
experiments, which, however, are often infeasible due to high costs or 
ethical concerns. In contrast, causal discovery aims to infer causal 
relations from observational data and learn the probabilistic graphical 
model of the underlying data. Historically, conventional causal discovery 
algorithms generally rely on carefully-crafted criteria to deduce graph 
structures. For instance, PC (Peter-Clark) algorithm conducts conditional 
independence tests to constrain graphical structures and gradually deduce 
the whole graph from data. As a result, they often produce spurious causal 
relations.

Recently, there is an emerging trend that seeks to use machine learning 
techniques to predict causal relations from observational data, instead of 
using hand-crafted criteria. These methods have achieved remarkable 
empirical performance compared to traditional methods. In this review, we 
discuss the theoretical foundations of supervised causal discovery (SCD) 
through the lens of learning theory and causal identifiability.

To show the impacts of causal discovery, we also present an application of 
SCD in data management and their distinct challenges beyond standard 
causal discovery. In particular, we introduce causality-based data 
explanations for interpreting query outcomes.

After providing thorough reviews of the theory foundations, empirical 
performance and applications of SCD, we discuss the research opportunities 
in SCD. We believe our study would benefit the causality community and 
data management community.


Date:  			Monday, 29 August 2022

Time:                  	3:00pm - 5:00pm

Zoom Meeting: 
https://hkust.zoom.us/j/96698667813?pwd=dWR6bjRKUVp6SjVGaitGbkl3VlUrQT09

Committee Members:	Dr. Shuai Wang (Supervisor)
 			Dr. Minhao Cheng (Chairperson)
 			Dr. Wei Wang
 			Prof. Raymond Wong


**** ALL are Welcome ****