Causal intervention is a practical approach for achieving unbiased (debiasing) and out-of-domain (OOD) generalization in classification. In recent years of visual causality research, the construction of causal graphs has often relied heavily on the researcher's own understanding of the problem, resulting in subjectivity and controversy. This report will explore: 1) how to objectively construct causal graphs, 2) the tangible benefits that such objective causal graphs can bring, and 3) based on some comparisons, summarize the current shortcomings of visual causal interventions.
Hanwang Zhang is an Associate Professor at Nanyang Technological University's School of Computer Science and Engineering. His research interests include Computer Vision, Natural Language Processing, Causal Inference, and their combinations. Due to his contribution in applied causality, he has received numerous awards including the Singapore President Award Young Scientist 2021, IEEE AI’s-10-To-Watch 2020, Alibaba Innovative Research Award 2019, Nanyang Assistant Professorship 2018, and several best paper awards.