Summary of Learning Cut Selection For Mixed-integer Linear Programming Via Hierarchical Sequence Model, by Zhihai Wang et al.
Learning Cut Selection for Mixed-Integer Linear Programming via Hierarchical Sequence Model
by Zhihai Wang, Xijun Li, Jie Wang, Yufei Kuang, Mingxuan Yuan, Jia Zeng, Yongdong Zhang, Feng Wu
First submitted to arxiv on: 1 Feb 2023
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Optimization and Control (math.OC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel hierarchical sequence model (HEM) is proposed for learning cut selection policies via reinforcement learning in mixed-integer linear programs (MILPs). The model tackles three key challenges: selecting the right cuts, determining the number of cuts to use, and ordering the selected cuts. By formulating the cut selection task as a sequence-to-sequence learning problem, HEM can learn effective heuristics for MILP solvers. Experimental results show that HEM improves efficiency in solving MILPs compared to human-designed and learning-based baselines on both synthetic and large-scale real-world datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Cutting planes are important for solving mixed-integer linear programs (MILPs), which are used to solve many real-world problems. The key challenge is selecting the right cuts, but there are two other important questions: how many cuts should be selected, and what order should they be in? Most current methods use rules that were designed by people, but machine learning can be used to learn better rules from a collection of MILPs. A new model called HEM (Hierarchical Sequence Model) is proposed to solve this problem using reinforcement learning. HEM learns the right number of cuts and their order simultaneously. Tests show that HEM works well on both small and large real-world problems. |
Keywords
» Artificial intelligence » Machine learning » Reinforcement learning » Sequence model
