Section: New Results

Space-efficient Acyclicity Constraints

Participant : Taus Brock-Nannestad.

Acyclicity constraints can be used to encode a large variety of useful constraints on graphs. The basic constraint itself can be encoded in terms of simpler constraints (e.g. integer linear constraints) in a straightforward and intuitive way, associating to each vertex of the (fixed) input graph a variable with domain linear in the size of the graph. For large graphs, this quickly becomes inefficient.

In [24], we show that in the case of planar graphs, a more efficient encoding (using a two-valued variable per vertex) is possible.