distillpub/post--bayesian-optimization

Website needs to be updated to new index.html

Closed this issue · 4 comments

The brackets in this formula are correct:

x_{t+1} = argmax(\alpha_{PI}(x)) = argmax(P(f(x) \geq (f(x^+) +\epsilon)))

But when displayed on the website, they are wrong: https://distill.pub/2020/bayesian-optimization/
image

Would also be good to run this article through a word processor with spelling/grammar checking as some of the English is a bit broken, for example:

In case of multiple points have the same <d-math>\alpha_{EI}</d-math>, we should prioritize the point with lesser risk (higher <d-math>\alpha_{PI}</d-math>). Similarly, when the risks are the same (same <d-math>\alpha_{PI}</d-math>) for any two points, we should choose the point with greater reward (higher <d-math>\alpha_{EI}</d-math>).

Is this better than before? It turns out a yes and a no. We see that here we do too much exploration, given the value of <d-math>\epsilon = 3</d-math>. We quickly reach close to global maxima, but unfortunately, do not exploit to get more gains near the global maxima.

Style of writing is very conversational, but might be intended.

Hi @colah, I updated the article 2 days [822b3fa] ago but the article has not updated on distill.pub yet. Can you please help me out with any steps that I might have missed?

Hi @jvanheugten,
Thanks a lot for creating an issue here with such rigor. We appreciate your help. :)

But when displayed on the website, they are wrong: https://distill.pub/2020/bayesian-optimization/

I think master is a protected branch, and therefore, there might be more steps before I can update the rendered article.

Would also be good to run this article through a word processor with spelling/grammar checking as some of the English is a bit broken, for example:

  • We have addressed the issues in this commit. [16e76f2]
  • Sure, we will run the article through a word processor and correct any issue it as able to find.

Closing this issue, as the article has been updated.