2017 IJCAI IJCAI 2017

On the Kernelization of Global Constraints

Abstract

Kernelization is a powerful concept from parameterized complexity theory that captures (a certain idea of) efficient polynomial-time preprocessing for hard decision problems. However, exploiting this technique in the context of constraint programming is challenging. Building on recent results for the VertexCover constraint, we introduce novel "loss-less" kernelization variants that are tailored for constraint propagation. We showcase the theoretical interest of our ideas on two constraints, VertexCover and EdgeDominatingSet.

🌉 Interdisciplinary Bridge — Artificial Intelligence and Computer Science and Machine Learning and Mathematics & Optimization
🧭 Keyword Pioneer — polynomial-time preprocessing
🐝 Cross-Pollinator — Artificial Intelligence, Computer Science, Data Science & Analytics, Deep Learning, Interdisciplinary, Knowledge & Reasoning, Machine Learning, Mathematics & Optimization, Natural Language Processing, Reinforcement Learning, Robotics