The $\infty$ sign mean in the energy formula might be a bit confusing, I would find it clearer to express the independent set constraint explicitly (unless the algorithm actually relies on this formulation with an objective barrier?)
The lattice graphs are really cool but it's hard to distinguish shades of gray alone, could we maybe vary edge thickness too?
The sampling step at the end returns an empty lattice
Thanks for the suggestions! We've updated the energy formula, fixed the empty lattice issue, and tweaked the lattice graph's edge thickness. It now looks like this:
https://github.com/openjournals/joss-reviews/issues/5700