There are a few students here at SciPy doing mathematical physics in Oslo. They seem to be interested in pursuing this idea. (One of them is CCed.)
They also suggested that Morten Hjorth-Jensen<http://folk.uio.no/mhjensen/cp/mhj.html>, faculty at Oslo, might be interested in funding some work on this sort of topic through their Computers in Science Education project. On Sat, Jun 29, 2013 at 6:03 AM, Stefan Krastanov < [email protected]> wrote: > I have done some work on this and I am very eager to do more if I get > some help. There are 3 (or 4) main parts in any such project: > > rules: > > 1. Given a Lagrangian, automatically deduce the Feynman rules (i.e. > the set of possible line types and the set of possible connections > between such lines). This is very well suited for > rewrite-rules/functional programing style. > > matrix elements (i.e. the values corresponding to each digram): > > 2. Given the Feynman rules, construct tree-level diagrams (a.k.a. > linear order diagrams / LO). Such diagrams do not contain loops and > are basically the first approximation of the nonlinear QFT process. > 3. Given the Feynman rules, construct diagrams with loops (a.k.a next > to linear order diagrams / NLO). Then comes the hard part: do all the > regularizations automatically. I do not really have the theoretical > background for this. > > sampling: > > 4. Actually evaluate the diagrams / matrix elements. This requires > sampling a high dimensional, very irregular space and contracting a > long product of tensors. Especially if we want realistic "cuts" (e.g. > do not count events in which the direction of the outgoing particle > does not coincide with a detector in the collider). > > > I have written a module for sympy that does 2. and generates fortran > code for 4. The code generation part was __very__ badly written, but > the rules->diagrams part was quite general. > > I would absolutely love to do steps 1 and 2 and contribute them to > sympy, but I will need help with that. First of all, we need to finish > the cleanup of Pernici's module for tensor canonicalization (needed > for 1.) and then add valued tensors (needed for 2.). My implementation > of valued tensors was hacked together and not appropriate for anything > more than a one-off use. > > So, if somebody is interested I would be very happy to organize an IRC > meeting so we can proceed on this. I wanted to do it for a while > (since 14 months ago), but did not have the time. > > For what I have done check out section 2 from the attached school > assignment paper. The title page is in French, but the rest is in > English. Do not laugh too much, it was implemented rather hastily :) > > The gist of the code is here: > > https://gitorious.org/school-lab-assignments/matrix-element-cruncher/blobs/master/v2/matrix_generator.py > Be warned, it is not reusable/extendable in its current state. > > > The leading projects for this in the physics community are: > > for 1. FeynRules which is a Mathematica package. It can do some very > fancy things, but it has some strange idiosyncrasies in the way it is > written. > > for 2. and 4. MadGraph5 > > for 3. there are no complete implementations > > On 29 June 2013 07:18, Matthew Rocklin <[email protected]> wrote: > > Has there been any work on implementing Feynman diagrams in sympy? Any > > thoughts on this topic? > > > > -- > > You received this message because you are subscribed to the Google Groups > > "sympy" group. > > To unsubscribe from this group and stop receiving emails from it, send an > > email to [email protected]. > > To post to this group, send email to [email protected]. > > Visit this group at http://groups.google.com/group/sympy. > > For more options, visit https://groups.google.com/groups/opt_out. > > > > > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sympy. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. For more options, visit https://groups.google.com/groups/opt_out.
