I have begun reading some of the (few) references I can find on the concept of a "core" of a non-negative matrix, e.g. here.
In brief, my question is as stated in the title. I would like to know what is an intuitive picture for the core of a matrix (as a limiting intersection of cones?)?
For some reason, this geometric idea has not seemed to gain much popularity in the community. It seems that it provides a way of understanding the Perron Frobenius theorem from a purely geometric perspective. This alone makes it seem like a useful idea, not to mention other insights that may be possible from this perspective.