Buu Phan

Papers

Total Citations

H-Index

About

Buu Phan is a researcher specializing in **Bayesian deep learning**, **uncertainty estimation**, and **computer vision**, with a particular focus on safety-critical applications. His work addresses one of the most pressing challenges in modern AI deployment: ensuring that machine learning systems can reliably quantify their own uncertainty, a crucial requirement in high-stakes environments such as autonomous driving and surgical robotics. Phan's most influential contribution, "Calibrating Uncertainties in Object Localization Task" (2018, 19 citations), advances the field of probabilistic object detection by developing methods that allow detection modules to output meaningful probability estimates alongside their predictions — enabling downstream systems to make safer, more informed decisions. Building on this foundation, his 2019 work "Bayesian Deep Learning and Uncertainty in Computer Vision" (11 citations) explores how intelligent robotic systems can better interpret visual data while accounting for the inherent risks of erroneous perception, directly linking uncertainty quantification to accident prevention. Together, these contributions position Phan as an emerging voice in the intersection of probabilistic machine learning and robotics safety — research areas of growing importance as autonomous systems become increasingly integrated into everyday life.