Skip to main content
12 events
when toggle format what by license comment
May 23, 2014 at 11:54 answer added Chassaing timeline score: 1
May 22, 2014 at 20:57 vote accept john mangual
May 21, 2014 at 14:52 history edited john mangual CC BY-SA 3.0
suggestion
May 21, 2014 at 14:41 answer added Chassaing timeline score: 5
May 21, 2014 at 12:52 history edited john mangual CC BY-SA 3.0
added 853 characters in body
May 21, 2014 at 5:21 comment added Anthony Quas Probably I'm telling you what you saw already. Apologies if this is the case.
May 21, 2014 at 5:20 comment added Anthony Quas Having a 1 min glance at the Granville article, it seems to me that the objects are integers and permutation. He factorizes integers into primes and gets a "partition of unity" from that (i.e. $(\log p_1/\log n,\ldots,\log p_d/\log n)$). He factorizes permutations into cycles and gets a partition of unity from that $(\ell_1(\sigma),\ldots,\ell_d(\sigma))$. So I think the analogy is 1) pick a prime of size roughly $e^N$ and find its partition of unity; pick a permutation on roughly $N$ symbols and find its partition of unity. Lo and Behold! they have (roughly) the same distribution!
May 20, 2014 at 23:58 answer added Qiaochu Yuan timeline score: 6
May 20, 2014 at 23:53 comment added john mangual I think the objects are $\mathbb{Z}/p\mathbb{Z}$ actions on the permutation groups $S_n$.
May 20, 2014 at 23:44 comment added Anthony Quas Do you think it's more than "if you compute certain statistics for (a) prime factorization; and (b) cycle decomposition of permutations, then for large $n$, the distributions are close to the same thing"?
May 20, 2014 at 23:32 history edited KConrad CC BY-SA 3.0
added 9 characters in body
May 20, 2014 at 23:18 history asked john mangual CC BY-SA 3.0