Statistical Foundations for AI IDEA Lab.
We approach intelligence from a statistical point of view.
We care about when learning is reliable, fair, and interpretable.
Our work connects theory and applications, starting from clear structural assumptions.
Recent Highlights
Previous slide
Next slide
Research Areas
Statistical Machine Learning
Deep Learning Theory
Fair AI
Explainable AI
Bayesian Statistics
Publications
Seokhun Park, Choeun Kim, Jihu Lee, Yunseop Shin, Insung Kong and Yongdai Kim (2026). Bayesian Neural Networks for Functional ANOVA Model. ICLR 2026.
Kunwoong Kim, Kyungseon Lee, Jihu Lee, Dongyoon Yang, Yongdai Kim (2026). Doubly-Regressing Approach for Subgroup Fairness. ICLR 2026.
Jihu Lee, Kunwoong Kim, Sehyun Park, Insung Kong, Dongyoon Yang, Yongdai Kim (2026). A Fair Bayesian Inference through Matched Gibbs Posterior. ICLR 2026.
Kyungseon Lee, Jongjin Lee, Insung Kong and Yongdai Kim. (2025). On the use of supervised anomaly detection algorithms for extremely imbalanced data. Journal of the Korean Statistical Society.
Hwichang Jeong, Insung Kong, Yongdai Kim. (2025). Learning deep generative models based on binomial log-likelihood. Neurocomputing, 651, 131009
Jinwon Park, Kunwoong Kim, Jihu Lee and Yongdai Kim. (2026) Fair Model-based Clustering. Accepted by AAAI 2026 Oral.
Lab Seminar
20260819
이해영
HAG: Hierarchical Demographic Tree-based Agent Generation for Topic-Adaptive Simulation
20260813
안영민
(NeurIPS 2024) Sigmoid Gating is More Sample Efficient than Softmax Gating in Mixture of Experts
한상문
(ICML 2025) Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based Learning
박세현
(ICLR 2026) DirMoE: Dirichlet-routed Mixture of Experts