Abstract
Overfitting refers to spurious correlations that inflate prediction outcomes and result in misleadingly optimistic models. We investigate the case of systematic overfitting occurring when changes in bandwidths, positions and line shapes are introduced to spectra with strongly overlapped bands and are modeled using popular matrix-decomposition techniques because these parameters amplify variance in steep spectral sections in a manner detached from any chemical or physical interpretations. Our analysis of the geometry of profile shapes demonstrates how indirect assessment of lateral spectral changes distorts original spectral information and creates unnecessary complexity that results in misleading interpretations. Alternative strategy is proposed that remedies this situation and aims to generally improve stability of spectral modeling in the future.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
