We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Graph-based calibration transfer

Abstract: The problem of transferring calibrations from a primary to a secondary instrument, i.e. calibration transfer (CT), has been a matter of considerable research in chemometrics over the past decades. Current state-of-the-art (SoA) methods like (piecewise) direct standardization perform well when suitable transfer standards are available. However, stable calibration standards that share similar (spectral) features with the calibration samples are not always available. Towards enabling CT with arbitrary calibration standards, we propose a novel CT technique that employs manifold regularization of the partial least squares (PLS) objective. In particular, our method enforces that calibration standards, measured on primary and secondary instruments, have (nearly) invariant projections in the latent variable space of the primary calibration model. Thereby, our approach implicitly removes inter-device variation in the predictive directions of X which is in contrast to most state-of-the-art techniques that employ explicit pre-processing of the input data. We test our approach on the well-known corn benchmark data set employing the NBS glass standard spectra for instrument standardization and compare the results with current SoA methods.
Comments: Preprint submitted to Journal of Chemometrics
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
DOI: 10.1002/cem.3319
Cite as: arXiv:2006.00089 [stat.ML]
  (or arXiv:2006.00089v1 [stat.ML] for this version)

Submission history

From: Ramin Nikzad-Langerodi Dr. [view email]
[v1] Fri, 29 May 2020 21:30:37 GMT (720kb,D)

Link back to: arXiv, form interface, contact.