New Weisfeiler-Leman features enhance algorithm selection in constraint programming

Researchers have developed a new method for algorithm selection in constraint programming by leveraging Weisfeiler-Leman (WL) features. This approach automates feature extraction, moving beyond traditional manual statistics to capture the underlying problem structure. The proposed cut-based representation, known as WLc, models structural partitions to provide a more nuanced predictive signal. Experiments using Support Vector Machines, Random Forests, and Multi-Layer Perceptrons on MiniZinc Challenge instances showed that WLc features outperformed existing methods. AI

IMPACT Introduces a novel automated feature extraction methodology for constraint programming, potentially improving efficiency and predictive accuracy.

RANK_REASON The cluster contains a research paper detailing a novel methodology for feature extraction in machine learning for constraint programming. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Constraint optimisation and landscapes
• constraint programming
• fzn2feat
• graph neural networks
• machine learning
• MiniZinc Challenges
• Multi-Layer Perceptrons
• random forest
• support vector machine
• Weisfeiler–Leman algorithm
• Xu et al 2018

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →