Skip to main content
13 events
when toggle format what by license comment
Dec 6, 2018 at 19:09 comment added Thomas Steinke @Cauchy I have a similar question, but in my case everything is independent. Do you know an answer for the independent case?
Oct 4, 2018 at 15:29 comment added Cauchy @SudeepRaja, not really. For dependent random variables the problem seems to be hard
Oct 2, 2018 at 5:48 comment added good bandit Cauchy, did you find the answer to this question? I'm working on a proof whose most important step needs a tighter inequality.
Oct 18, 2017 at 23:39 comment added Cauchy Yes it is, I am looking for specific upper bounds of the ratio of the KL divergences. Notice that is not always true for $f$ divergences and continuous random variables. Here are 2 papers related to this topic: arxiv.org/pdf/1411.3575.pdf arxiv.org/pdf/1508.06025.pdf.
Oct 18, 2017 at 13:03 comment added Ant Isn't this essentially the data processing inequality?
Jun 18, 2017 at 7:59 comment added Cauchy sorry I didn't know that
Jun 18, 2017 at 7:24 comment added S. Carnahan Since this is a duplicate of math.stackexchange.com/questions/2326512/… you should at least mention that you are asking in more than one place, to prevent unnecessary duplication of effort. That is a basic curtesy to people who are working for free.
Jun 18, 2017 at 5:17 comment added Cauchy it looks like it is a strict inequality ($P_1 \neq P_2 $), however, I was wondering if $P_3$ could participate to the right part.
Jun 18, 2017 at 1:48 comment added Igor Rivin Where does this question come from?
Jun 18, 2017 at 1:41 history edited Cauchy CC BY-SA 3.0
added 24 characters in body
Jun 18, 2017 at 1:31 history edited user44143 CC BY-SA 3.0
added reference
Jun 18, 2017 at 1:24 review First posts
Jun 18, 2017 at 7:10
Jun 18, 2017 at 1:23 history asked Cauchy CC BY-SA 3.0