Posts by Collection

portfolio

publications

Improving Generalization with Flat Hilbert Bayesian Inference

Published in International Conference on Machine Learning (ICML), 2025

We introduce Flat Hilbert Bayesian Inference (FHBI), an algorithm designed to enhance generalization in Bayesian inference. Our approach involves an iterative two-step procedure with an adversarial functional perturbation step and a functional descent step within a reproducing kernel Hilbert space.

talks

teaching

Mentor

High school course, ThaiBinh High School for Gifted Student, 2021

Courses: Functional Equation

Teaching Assistant

Undergraduate course, Hanoi University of Science and Technology, 2024

Courses: Introduction to Programming, Machine Learning