This repo contains complementary code for the paper Global Representation Ring and Knutson Index (authors: Dylan Johnston, Diego Martín Duro, Dmitriy Rumynin). arxiv: https://arxiv.org/abs/2403.18498
To run: Download and open the .ipynb files on a Jupyter notebook and run with a Sagemath kernel. We used Sagemath 9.8 but other versions should work also. The GAP package is required.
Note: The code was written in a Jupyter Notebook as this appeared to be the easiest way to use Sagemath on a Windows machine.
- Overview
- global_table.ipynb
- SameCharSameBurnside.ipynb
- SplittingLemma.ipynb
- BrauerPairs.ipynb
Let
The upshot is that all finite
Let
Note: The global representation ring (as in [Witherspoon, The Ring of Equivariant Vector Bundles on Finite Sets, 1995]) occurs when
It is well known that
There also exists a mark on such pairs. Namely, let
This information can be encoded in a table - with rows labelled by (simple
The final thing to note is that if any two elements
This repo contains code which investigates questions and properties of the reduced global tables.
This is the main script in the repo. The main function is 'reduced_global_table(G,N,print_it = True)'. You pass a GAP group
To compute the global table of
#global_table.ipynb script here
G = gap.SmallGroup(5,1)
N = gap.Subgroups(G)[1] #gap indexing starts from 1. The identity subgroup is always at index 1.
my_table = reduced_global_table(G,N,print_it = False)
#do something with the table
The character table and Burnside table of marks can be seen from a global table. Therefore, when looking for two groups with the same global table a good place to start is to find two groups with the same character table and Burnside table of marks. This is the purpose of this script.
There is a notion of a Knutson subindex of a
- A sequence
$1 \rightarrow N \rightarrow G \rightarrow \Gamma \rightarrow 1$ splits$\implies$ $1 \rightarrow N_p \rightarrow G_p \rightarrow \Gamma_p \rightarrow 1$ splits, where$(-)_p$ denotes taking Sylow p-subgroups (some care is needed here on which Sylow p-subgroups are taken, see the paper for details). - A sequence
$1 \rightarrow N \rightarrow G \rightarrow \Gamma \rightarrow 1$ splits$\implies$ $\mathcal{K}_r^S(G/N) = 1$ .
The backward implications are false, e.g. gap.SmallGroup(
Two groups are said to be a Brauer pair if they have the same character tables and power maps. This script aims to find Brauer pairs. This was used to investigate whether there is a link between groups having the same global table and being a Brauer pair. However, we later found that no such link exists.