We present a massively parallel Lagrange decomposition method for solving 0--1 integer linear programs occurring in structured prediction. We propose a new iterative update scheme for solving the Lagrangean dual and a perturbation technique for decoding primal solutions. For representing subproblems we follow Lange et al. (2021) and use binary decision diagrams (BDDs). Our primal and dual algorithms require little synchronization between subproblems and optimization over BDDs needs only elementary operations without complicated control flow. This allows us to exploit the parallelism offered by GPUs for all components of our method. We present experimental results on combinatorial problems from MAP inference for Markov Random Fields, quadratic assignment and cell tracking for developmental biology. Our highly parallel GPU implementation improves upon the running times of the algorithms from Lange et al. (2021) by up to an order of magnitude. In particular, we come close to or outperform some state-of-the-art specialized heuristics while being problem agnostic. Our implementation is available at https://github.com/LPMP/BDD.
翻译:我们提出了解决结构化预测中的0-1整型线性程序的大规模平行拉格朗格分解方法。 我们提出一个新的迭代更新方案,以解决拉格朗基双轨和用于解码原始解决方案的扰动技术。 为了代表子问题,我们遵循Lange等人(2021年),并使用二进制决定图(BDDs)。我们的原始和双重算法要求子问题之间很少同步,而对于BDDs的优化只需要简单的操作,而没有复杂的控制流程。这使我们能够利用GPU为我们的方法的所有组成部分提供的平行功能。我们介绍了MAP对Markov随机场的推断、四进制任务和开发生物学的细胞跟踪的组合问题实验结果。我们高度平行的GPU的实施在Lange等人(2021年)的算法运行时通过一个程度的顺序得到改善。特别是,我们接近或超越了某种状态的专业化高科技,而同时又有问题。我们在 https://github./comMPDDD。