This is a repo for a 2021 Bayesian deep learning workshop Uncertainty? Hands-on Bayesian neural networks with Tensorflow and Tensorflow Probability
The repository will be updated before the workshop. Stay tuned!
From medical research to small-data scenarios, whenever we want to understand how sure the model is about its own predictions, modeling uncertainty can be immensely helpful. During the workshop we’ll learn how to build Bayesian neural networks using Tensorflow and Tensorflow Probability to model uncertainty. At the end of the workshop, you’ll have practical knowledge how to create basic types of Bayesian neural network using Tensorflow ecosystem and you'll be able to apply these techniques to your own projects. To fully benefit from the workshop you need: - good practical knowledge of Python - practical understanding of deep learning principles - experience using Tensorflow (recommended) or other contemporary deep learning framework - good understanding of basic probability and basic distributions - familiarity with Bayes' theorem