Please read the new rule regarding the restriction on the use of AI tools. ×

Hyper Log Log Algorithm

Revision en2, by StellarSpecter, 2024-10-12 14:53:25

Hyper Log Log — link to wikipedia

I heard this is an important algorithm to learn to get to 1000 rating, Could anyone break it down in simple terms, or maybe share an analogy? Like, something a 900 rated would understand?

Q. What exactly is it doing? I mean, how does it "estimate" things instead of just counting them for real?

Q. Why does it use these "hash functions"? (Is it for randomness? Or...?)

Q. What is this "log-log" part in its name supposed to mean? Is there a double logarithm somewhere?

Q. How accurate is it, really? And how does it balance between accuracy and memory usage?

Thanks in advance, and sorry if my questions are noobish.

History

 
 
 
 
Revisions
 
 
  Rev. Lang. By When Δ Comment
en2 English StellarSpecter 2024-10-12 14:53:25 25
en1 English StellarSpecter 2024-10-12 14:52:59 746 Initial revision (published)