7 Bayesian learning

 

This chapter covers

  • Identifying extrapolation as the Achilles heel of DL
  • A gentle introduction to Bayesian modeling
  • The concept of model uncertainty, which is called epistemic uncertainty
  • The Bayesian approach as a state-of-the-art method to dealing with parameter uncertainty

This chapter introduces Bayesian models. Besides the likelihood approach, the Bayesian approach is the most important method to fit the parameters of a probabilistic model and to estimate the associated parameter uncertainty. The Bayesian modeling approach incorporates an additional kind of uncertainty, called epistemic uncertainty. You will see that incorporating epistemic uncertainty results in better prediction performance and, more appropriately, quantification of the uncertainty of the predicted outcome distribution. The epistemic uncertainty becomes especially important when applying prediction models to situations not seen during the training. In regression, this is known as extrapolation.

7.1 What’s wrong with non-Bayesian DL: The elephant in the room

7.2 The first encounter with a Bayesian approach

7.2.1 Bayesian model: The hacker’s way

7.2.2 What did we just do?

7.3 The Bayesian approach for probabilistic models

7.3.1 Training and prediction with a Bayesian model

7.3.2 A coin toss as a Hello World example for Bayesian models

7.3.3 Revisiting the Bayesian linear regression model

Summary

sitemap