Exact Combinatorial Optimization with Graph Convolutional Neural Networks

arXiv1906.01629MaRDI QIDQ6319975

Author name not available (Why is that?)

Publication date: 4 June 2019

Abstract: Combinatorial optimization problems are typically tackled by the branch-and-bound paradigm. We propose a new graph convolutional neural network model for learning branch-and-bound variable selection policies, which leverages the natural variable-constraint bipartite graph representation of mixed-integer linear programs. We train our model via imitation learning from the strong branching expert rule, and demonstrate on a series of hard problems that our approach produces policies that improve upon state-of-the-art machine-learning methods for branching and generalize to instances significantly larger than seen during training. Moreover, we improve for the first time over expert-designed branching rules implemented in a state-of-the-art solver on large problems. Code for reproducing all the experiments can be found at https://github.com/ds4dm/learn2branch.

Has companion code repository: https://github.com/whuang-io/Distributional_MIPLIB_eval

This page was built for publication: Exact Combinatorial Optimization with Graph Convolutional Neural Networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6319975)