Efficient Groebner bases and their application

45 views
Skip to first unread message

Zamrath Nizam

unread,
Mar 17, 2014, 2:27:01 AM3/17/14
to sy...@googlegroups.com

Hi everyone,

I would like to know a solid word on the current status of the Groebner basis
related improvements. As far as I am concern the project is still at the position
where these suggestions were proposed (looked at the source code in sympy
several times before draw this conclusion). Moreover the author of above
suggestions has done some other project related to SymPy in the same year.
This makes a confusion here. Therefore please be kind enough to give me a
definitive word on the current state to date.

The project ideas related to Groebner basis as I understood to be done in this
GSoC version is listed below. Kindly look at the project proposal related to
Groebner basis improvement of mine. If I have gone miles away from the project
that is actually needed, please correct me.

1. Implement a better selection criteria
        I prefer the "Normal strategy for F4" is more efficient than "Sugar". Any
        Suggestions?

2. Implement F4

3. Analyze which approach is better in what contexts
        Does this mean the pr should intelligently choose an approach over
        others with the different properties of polynomial sets?

4. Implement Groebner walk

5. Apply Groebner basis in integration of rational and transcendental functions
and simplification of rational expressions modulo a polynomial ideal
        This is of course not a trivial task as seen since the existing code should
        have a work around to have the Groebner basis to be utilized. As with past
        experience with work arounds, I would consider this would be bit of a
        challenging task. With the above tasks (1, 2, 3 & 4) being added, could I
        add this part as an additional task in to the project proposal? (Or if that is
        the case, then I would take up the challenge. Only concern is with the time)

And I am very disappointing to see that I had no single reply for my previous
post about this. And I still believe that this project is simply doable for any
GSoC candidate who has fascinating knowledge in linear algebra. And this
would be a breakthrough if an important tools like this is being improved to the
latest possible up to date efficiency.

Thank you.

Regards,
Zamrath Nizam

Reply all
Reply to author
Forward
0 new messages