Go Back > Extra Stuff > Blogorrhea > MattcAnderson

Thread Tools
Old 2017-03-18, 10:51   #1
MattcAnderson's Avatar
"Matthew Anderson"
Dec 2010
Oregon, USA

3×13×17 Posts
Default Goldbach's weak conjecture

Hi Mersenneforum,

The 3 primes problem, or weak Goldbach conjecture has been proved true by
Helfgott in 2013.

It simply states that all odd numbers < 7 are the sum of 3 odd prime numbers.


In My Humble Opinion (IMHO) this was quite a proof.

wanted you to know.

MattcAnderson is offline   Reply With Quote
Old 2017-03-18, 23:32   #2
CRGreathouse's Avatar
Aug 2006

22×1,493 Posts

Greater than 7, that is.

Yes, it was a big deal! The proof had three parts: major arcs, minor arcs, and verifying weak Goldbach on small values. The third part was joint work with Platt and was verified considerably further than necessary: to 8.875e30 when Helfgott needed only 10^27.

Here's my understanding of the major+minor arc decomposition. Instead of looking at the sum
of the number of ways a given odd number N can be written as a sum of three primes, Helfgott looks at the related sum
\sum_{p+q+r=N}\log p\log q\log r
which is a weighted version of the first which is easier to work with analytically.

Next the sum is transformed into an integral
\int_0^1 S(\alpha,x)^3 - \exp(-2\pi iN\alpha)d\alpha
which is then split into two pieces: the major arcs, which are small intervals around rationals with small denominators, and the minor arcs which are everything else.

Actually, the integral Helfgott actually uses is a somewhat more complicated 'smoothed' version of this integral, but the basic idea is the same: it can be broken into major and minor arcs, and proving that it takes on positive values for a given N shows that weak Goldbach holds for that N.

Helfgott spends a paper describing how to bound the minor arcs very precisely, then has a somewhat easier task (because of the groundwork he laid) on the major arcs. Combining the two Helfgott proves that the smoothed integral is at least 0.000422 * N^2 when N is an odd number greater than 10^27, which together with the numerical verification with Platt proves the weak Goldbach conjecture.

Helfgott cites a lot of software used in the proof including PARI, Maxima, Gnuplot, VNODE-LP, PROFIL / BIAS, SAGE, and Platt’s interval-arithmetic package (based in part on Crlibm).
CRGreathouse is offline   Reply With Quote

Thread Tools

Similar Threads
Thread Thread Starter Forum Replies Last Post
Factorial and Goldbach conjecture. MisterBitcoin MisterBitcoin 17 2018-01-29 00:50
Goldbach's weak conjecture MattcAnderson MattcAnderson 17 2017-03-21 18:17
Goldbach Conjecture MattcAnderson MattcAnderson 3 2017-03-17 15:34
Goldbach's Conjecture Patrick123 Miscellaneous Math 242 2011-03-15 14:28
Goldbach's conjecture Citrix Puzzles 3 2005-09-09 13:58

All times are UTC. The time now is 23:50.

Tue Mar 2 23:50:13 UTC 2021 up 89 days, 20:01, 0 users, load averages: 3.49, 2.75, 2.52

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.