Explore a 28-minute conference talk from the Uncertainty in Artificial Intelligence (UAI) 2023 Oral Session 2 that delves into the quantification of aleatoric and epistemic uncertainty in machine learning. Examine the appropriateness of conditional entropy and mutual information as measures for these uncertainties. Learn about the identified incoherencies in these information theory-based measures and the challenges surrounding the additive decomposition of total uncertainty. Gain insights from experimental results across various computer vision tasks that support the theoretical findings and raise concerns about current uncertainty quantification practices. Access the presentation slides to follow along with the speakers' arguments and visual aids.
Quantifying Aleatoric and Epistemic Uncertainty in Machine Learning - Session 2
Uncertainty in Artificial Intelligence via YouTube
Overview
Syllabus
UAI 2023 Oral Session 2: Quantifying Aleatoric and Epistemic Uncertainty in Machine Learning
Taught by
Uncertainty in Artificial Intelligence
Related Courses
-
Is the Volume of a Credal Set a Good Measure for Epistemic Uncertainty - UAI 2023 Oral Session 2
-
Human-in-the-Loop Mixup: Aligning Synthetic Data with Human Perception - Oral Session 2
-
Learning from Low Rank Tensor Data: A Random Tensor Theory Perspective - UAI 2023 Oral Session 4
-
Uncertainty Quantification with Physics-Informed Machine Learning
-
Uncertainty Quantification in Machine Learning: Federated Conformal Prediction
