Optimizing parameters of information-theoretic correlation measurement for multi-channel time-series datasets in gravitational-wave detectors

Abstract

Abstract Data analysis in modern science using extensive experimental and observational facilities, such as gravitational-wave detectors, is essential in the search for novel scientific discoveries. Accordingly, various techniques and mathematical principles have been designed and developed to date. A recently proposed approximate correlation method based on information theory has been widely adopted in science and engineering. Although the maximal information coefficient (MIC) method remains in the phase of improving its algorithm, it is particularly beneficial in identifying the correlations of multiple noise sources in gravitational-wave detectors including non-linear effects. This study investigates various prospects for determining MIC parameters to improve the reliability of handling multi-channel time-series data, reduce high computing costs, and propose a novel method of determining optimized parameter sets for identifying noise correlations in gravitational-wave data.