单目标优化景观的线性矩阵分解嵌入
Linear Matrix Factorization Embeddings for Single-objective Optimization Landscapes
摘要:
自动化的每实例算法选择和配置在许多经典优化问题上表现出了良好的性能,包括可满足性,AI规划和TSP。这些技术通常依赖于一组特征,这些特征测量手头问题实例的某些特征。在黑盒优化的上下文中,这些特征必须从一组$(x,f(x))$样本中导出。在文献中已经提出了许多不同的特征,例如,测量手头实例的模态、可分性或坚固性。然而,几个常用的特征是高度相关的。虽然最先进的机器学习技术可以常规地过滤这种相关性,但它们妨碍了所导出的算法设计技术的可解释性。因此,我们建议在这项工作中,通过表示学习预处理的测量(原始)景观功能。更确切地说,我们表明,通过矩阵分解的线性降维显着有助于更好地检测不同问题实例之间的相关性-成功的自动算法设计的关键先决条件。
英文摘要:
Automated per-instance algorithm selection and configuration have shown promising performances for a number of classic optimization problems, including satisfiability, AI planning, and TSP. The techniques often rely on a set of features that measure some characteristics of the problem instance at hand. In the context of black-box optimization, these features have to be derived from a set of $(x,f(x))$ samples. A number of different features have been proposed in the literature, measuring, for example, the modality, the separability, or the ruggedness of the instance at hand. Several of the commonly used features, however, are highly correlated. While state-of-the-art machine learning techniques can routinely filter such correlations, they hinder explainability of the derived algorithm design techniques. We therefore propose in this work to pre-process the measured (raw) landscape features through representation learning. More precisely, we show that a linear dimensionality reduction via matrix factorization significantly contributes towards a better detection of correlation between different problem instances -- a key prerequisite for successful automated algorithm design.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。