I could be off-base here, but I think they’re saying that a sparse matrix (for holding the edge weights) is ultimately something like a list-of-lists format, ra

I could be off-base here, but I think they’re saying that a sparse matrix (for holding the edge weights) is ultimately something like a list-of-lists format, rather than a dense n x m array (for which GPUs have optimized routines). Except I that there are libraries doing sparse matrix operations on GPU (https://developer.nvidia.com/cusparse https://developer.nvidia.com/cusparse).