Learning to branch with Tree-aware Branching Transformers

作者：

Highlights：

• A transformer-based framework (T-BranT) for branching is proposed.

• T-BranT captures the mutual connections between candidate variables.

• T-BranT integrates branching history which is conducive to branching.

• Experiments demonstrate that T-BranT achieves a significant boost on performance.

摘要

•A transformer-based framework (T-BranT) for branching is proposed.•T-BranT captures the mutual connections between candidate variables.•T-BranT integrates branching history which is conducive to branching.•Experiments demonstrate that T-BranT achieves a significant boost on performance.

论文关键词：Branch and Bound,Machine learning,Branching strategies,Mixed Integer Linear Programming

论文评审过程：Received 22 December 2021, Revised 27 May 2022, Accepted 12 July 2022, Available online 16 July 2022, Version of Record 19 July 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109455