Jungsoo Kim

Ph.D. student, SNU Causality Lab, Seoul National University

prof_pic.png

SNU Causality Lab

Graduate School of Data Science

Seoul National University

Seoul, Korea

I am a Ph.D. student in the Causality Lab at the Graduate School of Data Science, Seoul National University, advised by Prof. Sanghack Lee. My research interests include causal inference, representation learning, conformal prediction, and trustworthy machine learning. Recently, our paper, “A Tale of Two Uncertainties: Global-Local Attribution for Conformal Prediction,” which studies how global and local attribution can make the uncertainty behind conformal prediction more interpretable, was accepted as a spotlight paper at the EIML workshop at ICML 2026.

Before SNU, I studied Science, Technology and Policy and Quantitative Risk Management at Yonsei University’s Underwood International College, with an exchange semester in Mathematical and Computational Science at Yale-NUS College. Welcome anyone who has interest in questions that sit between statistical evidence, policy decisions, and real-world systems.

news

Jun 30, 2026 Our paper A Tale of Two Uncertainties: Global-Local Attribution for Conformal Prediction was selected as a spotlight paper at the ICML 2026 EIML workshop.

selected publications

  1. EIML
    A Tale of Two Uncertainties: Global–Local Attribution for Conformal Prediction
    Sangyeon Cho, Minyoung Cho, Jungsoo Kim, and 2 more authors
    In ICML 2026 Workshop on The 2nd Workshop on Epistemic Intelligence in Machine Learning: Learning under Unknown Unknowns for Real-world Impact, 2026
  2. CauScien
    Instrumental Variable Representation Learning under Confounded Covariates
    Jungsoo Kim, Kwonho Kim, Inwoo Hwang, and 1 more author
    In NeurIPS 2025 Workshop on CauScien: Uncovering Causality in Science, 2025