Mathematics desk
< July 29	<< Jun \| July \| Aug >>	Current desk >

Welcome to the Wikipedia Mathematics Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 30

General Term

Let ${\color {white}.}\quad a_{n}={\frac {(-1)^{n}}{\zeta (n)}}\cdot \int _{0}^{1}{\frac {\ln ^{n}x}{(1+x)^{2}}}dx.\quad {\color {white}.}$ Then the first few terms are:

a₂	a₃	a₄	a₅	a₆	a₇	a₈	a₉	a₁₀
1	⁠9/2⁠	21	⁠225/2⁠	⁠1395/2⁠	⁠19845/4⁠	40005	⁠722925/2⁠	⁠7243425/2⁠

I was not able to find this sequence at the OEIS, and I was wondering whether
someone might help me find a general formula or algorithm for generating it. —
86.125.202.236 (talk) 01:22, 30 July 2014 (UTC)[reply]

Hint:

{\frac {\ln ^{n}x}{(1+x)^{2}}}=\sum _{k=0}^{\infty }\ln ^{n}(x)(-1)^{k}(k+1)x^{k}

. If you work through the sums you should end up with

a_{n}=n!\cdot 2^{-n}\cdot (2^{n}-2)

. DTLHS (talk) 03:02, 30 July 2014 (UTC)[reply]

Resolved

– - Thank you! :-)

Interestingly, this sequence is intimately related to the enumeration of functions which map a discrete space to itself. RomanSpa (talk) 08:02, 31 July 2014 (UTC)[reply]

Spectrum Problem in FMT

So, mentioning 0-1 laws in an above question caused me to end up rereading [1] from Fagin, which made me get reinterested in spectrums. I noticed [2] from 2009, which is interesting. I was wondering if anyone had any other reading suggestions on the topic; or if any further significant progress has been made since that was published? --Also, it mentions that the twin primes (and a few other sets of primes) are spectrums, it mentions that this because each such set is rudimentary, and all rudimentary sets are spectra; however, I was curious if anyone was aware of what FO sentences they are spectra of? More generally, given a set S that is a spectrum, is it possible to work backwards and get a sentence it corresponds with - for example, finite fields give that prime powers are a spectrum, could we run this backwards knowing that prime powers are a spectrum? Thank you for any help:-) By the way, I notice that we have Spectrum of a theory, but this does not discuss the FMT notion of spectrum, nor anything relating to it - do we have a page that does, or is this something that is not yet covered by Wikipedia?Phoenixia1177 (talk) 05:16, 30 July 2014 (UTC)[reply]

FMT = Finite model theory. -- ToE 15:05, 31 July 2014 (UTC)[reply]

Some thoughts: I can't find anything specific to spectra in the sense of finite model theory described in your first link on WP, but perhaps I haven't looked hard enough. As you likely know, the word "spectrum" is semantically overloaded in math and physics, but there are usually some similarities. E.g. Spectrum_(functional_analysis) talks about the generalization of eigenvalues of bounded linear operators. ~~I'm unfamiliar with model theory, but there might be some connections there...~~ Edit: that is probably nonsense. My dim memories of Decomposition_of_spectrum_(functional_analysis)#Decomposing_the_spectrum and Essential_spectrum made them seem more similar to the FMT definition.

As for finding a sentence that 'generates' a certain spectrum, that sounds like an inverse problem, and those are notoriously ill posed. Like a good counter example, my hunch is that there is not a general purpose algorithm for constructing such sentences, it would seem to require a clever human familiar with the concepts. As for further reading, lacking any better suggestions, I would just skim through the works that cite the Fagin paper on Google scholar. So, no answers from me, but I thought the links might help. SemanticMantis (talk) 21:18, 31 July 2014 (UTC)[reply]