New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Question about random forest surrogate and high dimensional data #64
Comments
For your questions,
|
Thank you for the explanation, here is the paper I was talking about for the second point: About the third question, my understanding is that using RF instead of GP will partially solve the high-dimension problem, since GP is typically not good for high-dimension data. If we choose RF as the surrogate, what other problem we'll encounter when dealing with high dimensional data? |
Hi @wushanyun64, RF is better than GP in high-dimension problem. However, since data points are scarce compared to the high dimensionality of the space, traditional BO method tends to over explore the space and may behave like random search. Some high-dim methods tackle this problem by restricting the search in local area of existing points. |
Hi Openbox,
I have two quick questions that would really appreciate if you guys could help me better understand.
Thanks in advance,
The text was updated successfully, but these errors were encountered: