Google Summer of Code 2016

[Stefan van Zwam]

Hi all,

SageMath is putting together a proposal for GSoC 2016. Last year, Chao Xu did a great job implementing various connectivity and matroid intersection algorithms. Michael Welsh and myself were both mentoring him. I'd like to get another student working on graphs and matroids for this year's edition, so I'm all ears for ideas. What are SageMath's biggest shortcomings when it comes to matroid theory?

Here are the proposals from the past two years:

http://wiki.sagemath.org/GSoC/2015#Extending_Matroid_Theory_functionality

https://docs.google.com/document/d/15v7lXZR1U4H2pT21d2fyPduYGb74JAFjkXJ6CWYmYfw/edit?usp=sharing

So please share here:

1) project ideas/requests

2) whether you wish to mentor/co-mentor

The deadline for SageMath's application as a participating organization is FEBRUARY 19. The list of projects is an important part of this application, so please think about this before then.

Cheers,

Stefan.

[Michael Welsh]

> On 19/01/2016, at 0611, Stefan van Zwam <stefan...@gmail.com> wrote:
>
> 2) whether you wish to mentor/co-mentor

I'm up for this again.

[Stefan van Zwam]

Awesome! I added you to the project page as mentor. Any ideas on projects for this year?

On Monday, February 8, 2016 at 9:47:44 PM UTC-6, yomcat wrote:

[Stefan van Zwam]

By the way, here is this year's page:

https://wiki.sagemath.org/GSoC/2016#Extending_Matroid_Theory_functionality

[Rudi Pendavingh]

Hi Stefan,

I cannot commit to supervising the summer of code, unfortunately. But I do have some suggestions. First, looking at this year’s page:

https://wiki.sagemath.org/GSoC/2016#Extending_Matroid_Theory_functionality

3. I agree that certificates are a good idea, but if you want an isomorphism, you can already ask for M.isomorphism(N).

4. There already is a test for regular, binary, ternary in Sage. I coded these last summer. I made a ticket for quaternary too, but I ran out of time. Go ahead and do it, that would be a good project. 

5. I really like that one.

6. Peter Nelson has made a suggestion for testing isomorphism of (dense) binary matroids: count the number of intersections in the ambient projective space between lines spanned by pairs of points of the matroid, and use that as an invariant. 

I propose:

9. testing graphicness. This is low-hanging fruit. Cunningham’s algorithm for this is quite close to his 3-connectivity test, which has already been implemented in Sage. The existing algorithm works great on small matroids but has a poor asymptotic runing time. Cunningham’s algorithm would handle matroids up to 1000 elements gracefully, I estimate.

Best,

Rudi

--

---
You received this message because you are subscribed to the Google Groups "sage-matroid" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sage-matroid...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.