Visual comparison of parameter optimisation strategies

The task is to find a maximum of a custom function. Which method will win? And what are differences between different approaches? The workflows compares the new Bayesian optimisation and Random Search and previously available Brute search and Hillclimbing. Sampled points are coloured with a colour gradient from white (generated first) to red (generated last). Bayesian optimisation also shows a set of gray points- those are randomly sampled in the warm-up phase.


This is a companion discussion topic for the original entry at https://kni.me/w/w0ck8o1lFiYYpDa5