You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 31, 2023. It is now read-only.
Hello,
I'd like to apply LaMCTS to a discrete but highly dimensional search space (e.g., 200 dimensions).
Each dimension, however, has only four options.
Would LaMCTS be suitable for this? How hard is it to adapt the code for this?
The text was updated successfully, but these errors were encountered:
@thiagotei Hi, I'm working on a discrete search space problem as well but with less than 50 dimensions.
I just use a floor function to convert the continuous floating-point data to discrete interger data, and it works quite well.
I'm not sure if it can achieve good performance in your high-dimensional case, but I think it is not so hard to apply it on discrete problems.
Hope this helps.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Hello,
I'd like to apply LaMCTS to a discrete but highly dimensional search space (e.g., 200 dimensions).
Each dimension, however, has only four options.
Would LaMCTS be suitable for this? How hard is it to adapt the code for this?
The text was updated successfully, but these errors were encountered: