Acquisition Function

Creator
Creator
Seonglae Cho
Created
Created
2024 Oct 1 0:24
Editor
Edited
Edited
2025 Feb 4 14:45
Refs
Refs

Balancing exploration and exploitation by utilizing the model's mean (prediction) and uncertainty (variance)
 
 

Bayesian Optimization Process

  1. Choose a Surrogate model for modeling the true function  and define its prior over the space of objective functions to model our black-box function.
  1. Bayes Update: obtain or update Surrogate Posterior using Bayes’ rule by incorporating set of observations into the surrogate model
  1. Use Acquisition functions , which depend on the surrogate posterior, to determine next sample evaluation point
  1. Add newly sampled data to the set of observations and goto step #2 till convergence or budget elapses
 
 
 
 

Recommendations