Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Exact Combinatorial Optimization with Graph Convolutional Neural Networks

About

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.

Maxime Gasse, Didier Ch\'etelat, Nicola Ferroni, Laurent Charlin, Andrea Lodi• 2019

Related benchmarks

TaskDatasetResultRank
Mixed Integer Linear Programming SolvingCapacitated Facility Location 200x100 (Medium)
Nodes Explored362
16
BranchingCapacitated Facility Location Small
Time21
12
BranchingCapacitated Facility Location Large
Time438
12
Combinatorial OptimizationCombinatorial Auction Small s
Time1.23
12
Combinatorial OptimizationCombinatorial Auction Medium
Time12.9
12
Combinatorial OptimizationCombinatorial Auction Large
Computation Time142
12
Combinatorial OptimizationSet Covering Small s
Time (s)3.75
12
Maximum Independent SetMaximum Independent Set Small (s)
Execution Time3.65
12
Combinatorial OptimizationSet Covering Medium
Time15.27
12
Combinatorial OptimizationSet Covering Large l
Computation Time48
12
Showing 10 of 40 rows

Other info

Follow for update