Skip to main content
10 events
when toggle format what by license comment
Apr 19, 2016 at 20:06 comment added VSJ Another way to pose the question would be to take logarithms. The function $-\sum_{i = 1}^n x_i \log x_i$ is the popular entropy function of the distribution. The lower bound (which might still be true) then looks as follows: $H(\mathbf{x}) \geq 2\log \sum_{i=1}^n \sqrt x_i - \log 2.$ Perhaps some inequality like Pinker's might help to establish this.
Apr 12, 2016 at 16:37 vote accept Jennifer Gao
Apr 12, 2016 at 14:24 answer added Iosif Pinelis timeline score: 12
Apr 12, 2016 at 9:46 comment added martin cripps Sorry too many square roots in the last inequality - it should read $\sum_i\sqrt{x_i}\geq \prod_ix_i^{-2x_i}$!
Apr 12, 2016 at 9:27 comment added martin cripps If you use $\log\sum_i \sqrt{x_i} = \log\sum_i x_i \frac{1}{\sqrt{x_i}}\geq \sum_i x_i \log(\frac{1}{\sqrt{x_i}})=-2\sum_ix_i\log x_i$ (where the inequality is Jensen's) you get $\sum_i \sqrt{x_i}\geq \sqrt{\prod_i x_i^{-2x_i}} $. Which is enough for the lower bound, because $\prod_i x_i^{x_i}<1$.
Apr 12, 2016 at 9:22 comment added joro You assume $0^0=1$, right?
Apr 12, 2016 at 8:32 comment added Jennifer Gao @AndrejBauer Definitely a real problem, but it's a long story. I am hoping it is true because it allows me to describe the probability distribution of a sum of random variables whose cdfs have a certain form.
Apr 12, 2016 at 8:27 comment added Andrej Bauer Is this the sort of idel curiosity that lead to the discovery of so many things, or a question motivated by a "real problem"? (I am asking out of idle curiosity because I can't image how and why anyone would think of these inequalities.)
Apr 12, 2016 at 8:23 review First posts
Apr 12, 2016 at 8:39
Apr 12, 2016 at 8:22 history asked Jennifer Gao CC BY-SA 3.0