What question did this study set out to answer?

Develop a machine learning-guided branching strategy using graph neural networks to optimize exact branch-and-reduce algorithms for the NP-hard maximum independent set problem.

January 1, 2024Open Access

Targeted Branching for the Maximum Independent Set Problem Using Graph Neural Networks

View Full Paper

Ask AI

Bookmark

Discussion

Member takes

Implication

Algorithmic evaluation reveals graph neural network-guided branching accelerates maximum independent set solving across benchmark graphs, indicating improved efficiency for exact solvers.

Key Points

Develop a machine learning-guided branching strategy using graph neural networks to optimize exact branch-and-reduce algorithms for the NP-hard maximum independent set problem.
Designed a graph neural network (GNN) to determine vertex selection for branching during graph search.
Trained model parameters using a population-based genetic algorithm to bypass challenges associated with supervised and reinforcement learning in complex solvers.
Evaluated computational efficiency against conventional maximum-degree branching rules across benchmark graph instances.
Achieved a computational speedup on 73% of the evaluated benchmark graph instances.
Recorded an overall median solving time speedup of 24% compared to standard baseline branching heuristics.

Cite This Study

A 2024 study studied this question.

synapsesocial.com/papers/6a7c99d8fd7ae2306b770c92 https://doi.org/10.4230/lipics.sea.2024.20

View Full Paper

Ask AI

Bookmark

January 1, 2024Open Access

Targeted Branching for the Maximum Independent Set Problem Using Graph Neural Networks

View Full Paper

Ask AI

Bookmark

Discussion

Member takes

Implication

Algorithmic evaluation reveals graph neural network-guided branching accelerates maximum independent set solving across benchmark graphs, indicating improved efficiency for exact solvers.

Key Points

Develop a machine learning-guided branching strategy using graph neural networks to optimize exact branch-and-reduce algorithms for the NP-hard maximum independent set problem.
Designed a graph neural network (GNN) to determine vertex selection for branching during graph search.
Trained model parameters using a population-based genetic algorithm to bypass challenges associated with supervised and reinforcement learning in complex solvers.
Evaluated computational efficiency against conventional maximum-degree branching rules across benchmark graph instances.
Achieved a computational speedup on 73% of the evaluated benchmark graph instances.
Recorded an overall median solving time speedup of 24% compared to standard baseline branching heuristics.

Cite This Study

A 2024 study studied this question.

synapsesocial.com/papers/6a7c99d8fd7ae2306b770c92 https://doi.org/10.4230/lipics.sea.2024.20

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark