Efficient Message Passing for 0-1 ILPs with Binary Decision Diagrams

We present a message passing method for 0-1 integer linear programs. Our algorithm is based on a decomposition of the original problem into subproblems that are represented as binary decision diagrams. The resulting Lagrangean dual is solved iteratively by a series of efficient block coordinate ascent steps. Our method has linear iteration complexity in the size of the decomposition and can be effectively parallelized. The characteristics of our approach are desirable towards solving ever larger problems arising in structured prediction. We present experimental results on combinatorial problems from MAP inference for Markov Random Fields, quadratic assignment, discrete tomography and cell tracking for developmental biology and show promising performance.