Uncertainty is a key element in modelling the world around us. If we know how uncertain something is, we can make decisions about it.
Probabilistic models can make decisions given partial information about the world, account for noisy inputs, explain phenomena not part of our models, describe inherently stochastic behaviour in the world.
In Bayesian Machine Learning, the main idea is to define a model that expresses our knowledge about a problem with some unknown parameters. We can encode our prior beliefs about these parameters and update our beliefs as soon as we see new data. Finally, we can compute the posterior distribution of the parameters in order to make predictions and decisions.
The aim of the course is to provide the fundamental concepts of Bayesian Machine Learning, present the challenges and the computatonal issues related to the computation of the posterior distribution.
- Hands-on experience
- Team work
- Flipped classroom
- Open teaching material
- Have fun
- Learn to Model Uncertainty
