Samuel Wilson

I will get to work on a thorough benchmark against mlrMBO tomorrow (I haven't look too hard at the mlMBO package). I do have benchmarking available against the rBayesianOptimization package,...

I am actually working on a kernel now, unfortunately my time has been spent working on improvements to the package instead of benchmarking. Here is an example of a kernel...

The API has changed after release 1.1.0 - BayesianOptimization() has been replaced with bayesOpt() and its supporting methods. See the readme for examples. On Sun, Jan 10, 2021 at 2:51...

Hmmm this is an interesting suggestion. I know how I would solve it - just put constraints on the acquisition function. However, this is not possible as of now. I'll...

It is not currently possible - right now the process searches anywhere within the parameter bounds for the acquisition function optimum. I looked into this a little bit and it...

This package is overdue for an update in the parallelization. I need to do some research though. The way the package is set up, custom parallelization schemas are not supported,...

The search for the acquisition function optimum is run in parallel, see [getLocalOptimums](https://github.com/AnotherSamWilson/ParBayesianOptimization/blob/master/R/getLocalOptimums.R). Is this what you are referring to?

Yes this package has this ability. See this section of the [readme](https://github.com/AnotherSamWilson/ParBayesianOptimization#sampling-multiple-promising-points-at-once). In fact this package and the multi-point implementation, pre-dates that paper by a few months, lol.

Ahhh it's not as simple as I thought, this looks pretty cool. Do you know of any implementations, in any language? It looks pretty tough to implement.

We definitely want to keep the ability to run the scoring function in parallel - not all users of the package are tuning hyperparameters that allow multithreading. One option would...