Bayesian optimization: Difference between revisions

Content deleted Content added
Tags: Mobile edit Mobile web edit Advanced mobile edit
Dwiddows (talk | contribs)
Added wikilink to exploration and exploitation and removed dangling wikilinks from other descriptions that don't have links (and maybe shouldn't because some are very basic ideas)
Line 26:
==Acquisition functions==
Examples of acquisition functions include
* [[probability of improvement]]
* [[expected improvement]]
* [[Bayesian expected losses]]
* [[upper confidence bound]]sbounds (UCB) or [[lower confidence bound]]sbounds
* [[Thompson sampling]]
and hybrids of these.<ref>Matthew W. Hoffman, Eric Brochu, [[Nando de Freitas]]: Portfolio Allocation for Bayesian Optimization. Uncertainty in Artificial Intelligence: 327–336 (2011)</ref> They all trade-off [[Exploration-exploitation dilemma|exploration and exploitation]] so as to minimize the number of function queries. As such, Bayesian optimization is well suited for functions that are expensive to evaluate.
 
==Solution methods==