Timeline for probabilities of increasing events under different product measures.
Current License: CC BY-SA 2.5
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 5, 2010 at 10:56 | vote | accept | James Martin | ||
| Jun 3, 2010 at 19:35 | answer | added | James Martin | timeline score: 1 | |
| Jun 2, 2010 at 4:05 | answer | added | Ryan O'Donnell | timeline score: 3 | |
| May 31, 2010 at 21:25 | answer | added | Alekk | timeline score: 0 | |
| May 31, 2010 at 20:59 | comment | added | Fedor Petrov | we may think on second measure $P^{+}$ as a coupling of $P^{-}$ and another Bernoulli scheme (that is, $B^{+}=\max(B^{-},C)$, where $C$ takes values 0,1 with suitable probabilities). In this case, any element of $cn$-neighborhood of $A=\{(B_1,\dots,B_n)\}$ w.r.t Hamming distance would hold with probabilty almost 1, provided $c$ is small enough. So, your question essentially reduces to isoperimetric inequalities on Bernoulli cube. Sorry that it is not answer, and is not very clear | |
| May 31, 2010 at 18:04 | history | asked | James Martin | CC BY-SA 2.5 |