Uncertainty in Deep Learning

Eligible: Undergraduate and Masters students

Mentor: Paul Pu Liang

Description: Quantifying what neural networks don’t know and when they should abstain from making predictions is an important goal for safe real-world decision-making. This project will involve designing algorithms that can quantify uncertainty in neural networks and explore their applications towards noisy labels, outlier detection, interpretability, and robustness.

Skills/Experience: Prior experience in deep learning is an advantage but not a requirement.

Contact: Interested students should send an email to Paul Pu Liang with their CV and description of their experience in machine learning.