Insung Kong

Insung Kong is a researcher specializing in machine learning and statistics, with a focus on theoretical foundations. His work addresses key challenges in algorithmic fairness, representation learning, and neural network theory. In 2025, he has published five research articles in prestigious venues including Transactions on Machine Learning Research and IEEE Transactions on Pattern Analysis and Machine Intelligence.

His primary research areas include machine learning, statistical learning theory, probability theory, and optimization theory. He investigates topics such as fairness in AI, expressivity of deep networks, nonparametric regression, and integral probability metrics. His approach combines rigorous mathematical analysis with practical machine learning applications, particularly in developing theoretically grounded fairness frameworks and analyzing neural network architectures.

Kong's recent publications reveal a strong emphasis on fairness and theoretical guarantees in machine learning. He has developed novel methods for fair representation learning using integral probability metrics and analyzed the expressivity of deep Heaviside networks. His work also explores tensor product neural networks for functional ANOVA models, contributing to interpretable machine learning while establishing convergence rates and VC dimension bounds.