Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

I'd like to learn about modular forms. My background is mostly computational algebra and group theory, and I've had little-to-no training in complex analysis. I've briefly seen modular forms in a short literature review I did on Monstrous Moonshine. I've been scouting out various books, and most have a reasonably strong analytic focus.

While I'm happy to learn enough analysis to get into modular forms via the standard recommended textbooks, I was wondering:

Is there a good reference for learning modular forms that had a particularly algebraic or computational bent?

share|improve this question
6  
How about Modular Forms: A Computational Approach, by William A. Stein ? It is available on his home page at williamstein.org. –  Chandan Singh Dalawat Jan 12 '12 at 11:59
1  
Another good reference is the springer book \textit{A first course in modular forms} of Diamond and Shurman. Or if you wish something more algebraic, you can have a look at the paper of Diamond and Im entitled \textit{Modular forms and modular curves}. –  Nicolas B. Jan 12 '12 at 13:21
    
What do you mean by "group theory" in your context? If this includes things like Lie groups and algebraic groups, then there is another nice approach to modular forms and a much wider class of objects called automorphic forms. Analysis still plays a role, but the approach has a very different flavor that might appeal to you if you're comfortable with stuff like the structure theory of reductive groups and such. Dan Bump's book on automorphic forms is a very readable introduction to this approach if you've had some experience with classical modular forms. –  Ramsey Jan 12 '12 at 15:21
    
@Chandan Singh Dalawat: Oh nice! I was unaware there was a free version. –  BrettW Jan 12 '12 at 21:33
    
@Ramsey: In my context, finite group theory. Specifically p-groups and the classification of finite simple groups. I've had exposure to Lie groups and algebraic groups, so Dan Bump's book also sounds good. –  BrettW Jan 12 '12 at 21:35
add comment

1 Answer 1

up vote 7 down vote accepted

You can do a great deal with no analysis whatsoever, by defining modular forms of weight $k$ to be sections of the line bundle $\omega^{\otimes k}$ over the elliptic moduli stack. That sounds quite scary, but it can be made very elementary and concrete after a couple of pages of preparatory discussion. Deligne's "Courbes elliptiques: formulaire" is a good place to look, and quite a lot of that paper is also discussed in Appendix B to "Elliptic spectra, the Witten genus and the Theorem of the Cube" by Mike Hopkins, Matthew Ando and myself. Note that this approach gives the ring $$ MF_\ast = \mathbb{Z}[c_4,c_6,\Delta]/(1728\Delta-c_4^3+c_6^2) $$ of modular forms over the integers, not over $\mathbb{C}$. However, if you are interested in Moonshine you may want to construct the $q$-expansion homomorphism $MF_\ast\to\mathbb{Z}[[q]]$. I don't know a fully satisfactory treatment of that without using any analysis.

share|improve this answer
3  
A treatment can be given using the Tate Curve, which you propably know. That is an elliptic curve $y^2 + + xy = x^3 + a_4 x + a_6$ over the ring $\mathbb{Z}[[q]]$ with coefficients $a_6, a_6$ multiples of the Eisensteinseries $E_4$ and $E_6$. It carries a canonical nowhere vanishing differential (as every affine elliptic curve) so you can evaluated a modular form on it and obtain an element in $\mathbb{Z}[[q]]$. –  Thomas Nikolaus Jan 12 '12 at 20:07
1  
Moonshine will at some point require consideration of level greater than one, and poles at cusps. Also, it might be worth pointing out that the $\omega$ you use is not the usual sheaf of differentials. –  S. Carnahan Jan 13 '12 at 3:34
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.