
Yarin Gal
Professor · Department of Computer Science · University of Oxford
At Prof Rod’s reading desk
Questions to explore
- When should uncertainty change a system’s decision?
- What assumptions connect a practical neural network to a probabilistic model?
Research interests
Yarin Gal studies uncertainty, Bayesian deep learning and reliable machine learning.
- Uncertainty
- Bayesian deep learning
- Reliable machine learning
Explore the work
Prof Rod’s reading notes
Start with a paper
ICML · 2016
Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
Prof Rod’s reading note: Read the assumptions behind treating dropout as approximate Bayesian inference. Ask where uncertainty estimates could fail outside the training distribution.
Bring a question of your own
Turn reading into research
Write one decision where uncertainty would matter. Separate uncertainty in a prediction from variation caused by repeated evaluation runs.
Prof Rod’s evaluation lesson helps you ask what was sampled and what a comparison supports. This is our suggested preparation, not the researcher’s admission requirement.
Learn to compare model resultsKeep exploring
More at University of Oxford
Explore the universityImprove this page
Suggest a correction, share a paper to start with, or point readers to a group page or opening.