Archive for the ‘arxiv’ Category

Some ArXiv stats for 2017

December 29, 2017

[Posted on december 29, 2017.]

Since the last batch of preprints on the arXiv got out today (european time, at least) here are a few things I’ve noticed in 2017.

The 5th digit for monthly paper numbers was used several times, the precise numbers being : january=9186, february=8910,  march=11008, april=9029, may=11194, june=10297, july=9980, august=9854, september=10517, october=11627, november=11589, december=10011.

That’s a grand total of 123,202 papers in 2017. Sadly the number of different authors is not easily obtainable.

There were 113,380 submissions in 2016, so it’s an 8.66% increase this year. The increase from 2015 to 2016 had been of 7.69%, but since two new sections (Economics, and Electrical engineering & Systems Science) were started last september, that’s not directly comparable.

See also this page for an official graph of monthly submissions starting in 1994, while monthly download rates are equally fascinating (note the spike this very december!).

Assuming individual authors between 2015 and 2017 had on average a stable number of yearly submissions (of course giving, for a given author, a weight of 1/n to his papers with n-1 coauthors), this would imply an increase of about 8% new authors both in 2016 and 2017.

Who are they? Surely new PhD students for some part, but there are also folks who stopped publishing after obtaining their PhD two years ago, so the 8% figure is a balance between those two populations.  This should be compared to an in-depth count of PhD offers worldwide over the past few years to confirm this scenario, but I don’t have the time to do it.

Should this fail (i.e. offers of PhD didn’t grow much between 2014 and 2016) then probably the initial assumption is incorrect, which would mean authors in fact are publishing more on average. I do wonder which alternative is the correct one…

[Edit (january 3): the official 2017 stats are now out.]

 Winter solstice sunset over Antler Peak,

by Yellowstone National Park on flickr

Advertisements

Remarks on ζ(2n+1)

December 15, 2016

[Posted on december 15, 2016.]

Having for a few days stupidly and incorrectly claimed a proof of the irrationality of all \zeta(2n+1) after getting confused with a definition (sic) for which I express my deepest apologies for any confusion this may have caused, a new version of the preprint has withdrawn that claim.

The little that survives then is the formula \displaystyle \frac{\zeta(2n+1)}{r^{2n}}=(-1)^{n+1} \int\limits_{[0;1]^n} \left (\prod_{i=1}^{n}\frac{\log(x_i)}{x_i}\right ) \log\left (1-\left(\prod_{i=1}^nx_i\right)^r \right ) dx_1\cdots dx_n which for r=1, due to its product form, I thought might have some uses that the classical \displaystyle \zeta(k)=\int\limits_{[0;1]^k} \frac{dx_1\cdots dx_k}{1-x_1\cdots x_k} might not have.

Related expressions that I then obtained, namely \displaystyle \int\limits_{[0;1]^2} \frac{\log(x)}{x}\frac{\log(y)}{y}\log(1-(xy)^1) \log(x)\log(y) (xy)^{2k+1} dxdy =
\displaystyle  \frac{4\zeta(6)}{(2k+1)} + \frac{4\zeta(5)}{(2k+1)^2} + \frac{4\zeta(4)}{(2k+1)^3} + \frac{4\zeta(3)}{(2k+1)^4} + \frac{4\zeta(2)}{(2k+1)^5}  -\frac{4}{(2k+1)^6}\sum_{n=1}^{2k+1}\frac{1}{n}-4\sum_{j=1}^{6}\left (\frac{1}{(2k+1)^j} \sum_{i=1}^{2k+1}\frac{1}{i^{7-j}}\right )

and

\displaystyle \int\limits_{[0;1]^2} \frac{\log(x)}{x}\frac{\log(y)}{y}\log(1-(xy)^2) \log(x)\log(y) (xy)^{2k+1} dxdy =
\displaystyle  \frac{63\zeta(6)}{8(2k+1)} + \frac{31\zeta(5)}{4(2k+1)^2} + \frac{15\zeta(4)}{2(2k+1)^3} + \frac{7\zeta(3)}{(2k+1)^4} + \frac{6\zeta(2)}{(2k+1)^5}
\displaystyle  +\frac{8\log(2)}{(2k+1)^6}-\frac{4}{(2k+1)^6}\sum_{i=1}^{k}\frac{1}{2i+1}-8\sum_{j=1}^{6}\left (\frac{1}{(2k+1)^j}\sum_{i=0}^{k}\frac{1}{(2i+1)^{7-j}}\right )

do not seem to be of any help towards a proof of irrationality of at least \zeta(5).  A combination of those two expressions, perhaps with other expressions such as those of Vasilyev, might allow to improve matters, but I was not able to find how.

 

 

First quick look at Semantic Scholar

November 14, 2016

[Posted on november 14, 2016]

Since the arXiv has announced today that it is teaming up with Semantic Scholar for a year (a search engine that has received some publicity lately), let’s have a first quick look.

What does it say for Jean-Pierre Serre ? Only two “influential publications” to his name, this is surely ridiculous. Wait, the earliest date of publication is set to 1989, and I can’t find a way to change it…

Let’s pick some recent authors then: Terence Tao. Now the default range is 1978-2016, and the 6th most relevant paper is “Professor Terence Tao Visit – 27 August Canterbury Statistics Open Day” by Jenifer Brown.  Not too convincing either.

One more try: Cedric Villani (to simulate someone typing with a qwerty keyboard without accents). Just one page of results. What difference an accent aigu makes. The suggestions in the toolbar only proposed some Cedrics and no Cédric.

I may have missed some obvious settings, but if not I don’t think it is very useful yet.

***

In other news:

 

Early september items

September 4, 2016

[Posted on september 4, 2016.]

An assortment of recent short news items :

20160707_200809

Metz, a farewell. (August  2016, Public Domain).

Further august 2016 news

August 18, 2016

[Posted on august 18, 2016.]

In no particular order :

  • there is a website to gather comments on the forthcoming MSC 2020 (one obvious set of candidates not yet mentionned are the various perfectoid structures–Peter Scholze had used a bunch of MSC 2010 classes, but with more than 120 citations to that paper by now it is clear new ones are needed)
  • statistics papers in Inventiones are very rare, so this one must be outstanding
  • Cédric Villani will give a public lecture in Mumbai tomorrow
  • a student who ranked joint first at this year’s highly difficult competitive exam to enter l’X credits the book What is mathematics (2nd edition) by Courant, Robbins and Stewart, which he read while in eleventh grade (première), for giving him a passion for maths
  • also joint first was Cécile Gachet, who equally ranked first at the even more difficult competitive exam to enter ÉNS Ulm (this feat was obviously saluted, e.g. on twitter)
  • the MSRI program on Geometric Group Theory has just started, with in particular an introductory workshop next week
  • there’s an interesting article on special values of Zeta functions in the september issue of Notices of the AMS (including \zeta(2) as the volume of a moduli space)
  • topically, this blog’s host has attempted to come up with something at least remotely interesting a few weeks ago, it didn’t go so well as usual, but hopefully the first section is indeed new and worthy of publication, we shall see…

20160813_072506Louis XVIth, the seagull, and the antiparallelogram at dawn.

Nantes, august 2016 (public domain).

Further july items

July 9, 2016

[Posted on july 9, 2016.]

Spotted recently :

  • the results of the arXiv survey (previously discussed on this blog) are out and interesting (in particular the share of respondents below 30 is higher than I would have expected), hopefully the new search tools in particular will be ready by the end of the year
  • by the way, the extra digit in the identifier still hasn’t been needed, but it is getting close reaching 09644 last month after a record of 09792 in may
  • the paper that Shinichi Mochizuki produced last year has now been published, and there’s a rumor by David Hansen that the 4 original ones have been accepted
  • the MSRI summer school on the McKay conjecture starts next week
  • the youtube channel of CIRM has a playlist dedicated to talks related to the forthcoming 7th ECM,   while the channel of IHÉS has one for the june conference of its Trimestre Ondes Non Linéaires
  • the 2016 IMO takes place next week, maybe an occasion for a new mini-polymath
  • as it happens, the coldest place in the solar system is named after Charles Hermite

20160707_182306Temporary cubes next to Centre Pompidou-Metz

(july 2016, Public domain)

That arXiv questionnaire, and other news

April 16, 2016

[Posted on april 16, 2016]

Filling that arXiv questionnaire (now offline) was interesting. No idea how many people answered, and more importantly how biased that sample will turn out to be. I hope, for the sake of transparency, that they’ll quickly make the numbers freely available (the free-comment sections are of course private).

Also, it’s informative to see how varied opinions can be. I do agree with some points made by Izabella Laba in her blog post : no comments please (think low quality MO questions that quickly and deservedly get many downvotes, or the sometimes very irrelevant comments made by amateurs on blogs). That would surely drive lots of serious folks away. Some people agree, others don’t. On the other hand,  flagging “substantial overlap” could be useful if properly defined, IMHO.

As for citations stats and tools, taking a well-known preprint that never got formally published, by just clicking on the NASA ADS link one easily gets useful citation tools, while the blog trackbacks are suitably moderated. Not sure what could be added on top of that.

In other news :

  • topically, a math.GM paper on Navier-Stokes made it to a local story
  • a wonderfully clear and interesting talk by Mireille Bousquet-Mélou at CIRM (in french, but with slides in english) on plane lattice paths avoiding a quadrant, a topic related to a series of works done in the past 15 years by lots of people (and where many nice things occur, like the issue of finiteness or not of a certain group naturally associated to the path counting method)

  • an interview of Manjul Bhargava in CNRS News made after the conference mentioned in the previous post
  • a job ad for a mathematician in the gaming industry in Dublin

arXiv identifiers to be extended by one digit next january

December 21, 2014

So says the what’s new page. How close have they come to yymm.9999 to warrant this?

Well, in september, the post-holidays peak, it got to 1409.8676 (found this by dichotomy, maybe there’s a more clever way).  Actually, in october it went even higher to 1410.8871, while in november it receded a bit to 1411.8006.

All this is a bit too close for comfort, indeed.

Edit: and the year finished with 1412.8770

Giuga-Agoh redux

June 9, 2011

Some years ago I mentionned that conjecture, but didn’t really make any progress then.  Recently, I’ve tried again and found what could be a simple proof, which I’ve put here.

There may still be a silly mistake somewhere, but at least that seems to be going in the right direction. Comments welcome…

UPDATE (9 june 2011): well, it was blatantly wrong indeed…  The paper has now been retracted, with apologies for the waste of time.

 

an Arxiv rare event

June 16, 2010

In the past week, the math arxiv has seen both a preprint dated june 7  by Akhil Mathew, probably one of the youngest posters there (if not the youngest), and another dated june 12 by Jean-pierre Serre, probably one of the oldest posters there (if not the oldest). Now that’s noticeable!  🙂