Suppose now I have 1000 d-dimensional data points. Is there a convenient algorithm that can approximate the volume of the concave/convex hull formed by these points?
Thanks in advance!
Suppose now I have 1000 d-dimensional data points. Is there a convenient algorithm that can approximate the volume of the concave/convex hull formed by these points?
Thanks in advance!
Copyright © 2021 JogjaFile Inc.
According to Wikipedia, given a membership oracle, we can compute an $\epsilon$ approximation of the volume of the convex hull in polynomial time in $n$ (=1000), $d$ and $\frac{1}{\epsilon}$.
Combining with this answer, it seems that membership can also be tested fairly quickly. However, if this is convenient enough for you, I don't know.