arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2105.13645math.OCcs.LG
收藏

学习有效混合整数规划的割集选择

Learning to Select Cuts for Efficient Mixed-Integer Programming

Shanghai Jiao Tong University(上海交通大学);Noah’s Ark Lab, Huawei Technologies(华为诺亚方舟实验室);University College London(伦敦大学学院)

Zeren Huang, Kerong Wang, Furui Liu, Hui-ling Zhen, Weinan Zhang, Mingxuan Yuan, Jianye Hao, Yong Yu, Jun Wang

摘要:

割平面法在求解混合整数规划(MIP)问题的现代求解器中起着重要的作用。适当选择割集可以在早期阶段去除不可行解,从而在不影响解的精度的情况下大大减少计算负担。然而,主要的切割选择方法严重依赖于遗传算法,这在很大程度上取决于手头的具体问题,从而限制了它们的泛化能力。在本文中,我们提出了一个数据驱动的和可推广的切割选择方法,命名为切割排名,在多实例学习的设置。为了衡量候选切割的质量,一个评分函数,它采取的实例特定的切割功能作为输入,训练和应用在切割排名和选择。为了评估我们的方法,我们在合成数据集和真实世界的数据集上进行了广泛的实验。与通常使用的切割选择算法相比,基于学习的策略被证明是更有效的,并且能够推广到具有不同属性的多个问题。切割排名已部署在一个工业解决方案的大规模MIP。在每天多个$10^7$变量和约束条件的产品规划问题的在线A/B测试中,Cut Ranking在不损失解的精度的情况下,实现了比生产求解器平均12.42%的加速比。

英文摘要:

Cutting plane methods play a significant role in modern solvers for tackling mixed-integer programming (MIP) problems. Proper selection of cuts would remove infeasible solutions in the early stage, thus largely reducing the computational burden without hurting the solution accuracy. However, the major cut selection approaches heavily rely on heuristics, which strongly depend on the specific problem at hand and thus limit their generalization capability. In this paper, we propose a data-driven and generalizable cut selection approach, named Cut Ranking, in the settings of multiple instance learning. To measure the quality of the candidate cuts, a scoring function, which takes the instance-specific cut features as inputs, is trained and applied in cut ranking and selection. In order to evaluate our method, we conduct extensive experiments on both synthetic datasets and real-world datasets. Compared with commonly used heuristics for cut selection, the learning-based policy has shown to be more effective, and is capable of generalizing over multiple problems with different properties. Cut Ranking has been deployed in an industrial solver for large-scale MIPs. In the online A/B testing of the product planning problems with more than $10^7$ variables and constraints daily, Cut Ranking has achieved the average speedup ratio of 12.42% over the production solver without any accuracy loss of solution.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑