The Bayesian inference process:

1. Specify a prior probability distribution reflecting our knowledge of the problem.

2. Collect Data.

3. Calculate the Likelihood: how well the observed data aligns with each possible hypothesis or model.

4. Apply Bayes’ Theorem: Bayes’ theorem is used to update the prior probabilities into posterior probabilities, considering the data and the likelihood.

5. Make Informed Decisions: The posterior probabilities provide updated beliefs about the hypotheses, models, or parameters. We can use these posterior probabilities to make decisions, make predictions, or perform further analysis.

Here, we study robust decision-making in machine learning algorithms using Bayesian Inference.

Reference: Understanding bayesian inference