CNNs for Vis-NIR Chemometrics: From Contradiction to Conditional Design

TL;DR

This paper reviews the inconsistent conclusions in Vis-NIR CNN chemometrics studies, attributing them to uncontrolled variables, and proposes a physics-aware, conditional design framework for better model comparison.

Contribution

It introduces a conditional design framework linking CNN choices to spectral physics, dataset regimes, and deployment scenarios to improve reproducibility.

Findings

Disagreements stem from measurement complexity and validation design.

Effective receptive field mismatch affects spectral structure capture.

Hyperparameter tuning and validation strategies influence model rankings.

Abstract

Near-infrared (NIR; a.k.a.\ NIRS) deep-learning studies in chemometrics increasingly report mutually inconsistent conclusions regarding convolutional neural network (CNN) design, including small versus large kernels, shallow versus deep architectures, raw spectra versus preprocessing, and single-domain training versus transfer learning. As a result, the same architecture can appear superior in one study and inferior in another, creating a practical impasse for chemometric practitioners. In this review, we argue that these contradictions are not evidence of irreconcilable methods but a structurally expected consequence of uncontrolled moderating variables. Specifically, we trace recurring disagreements to (i) the indirect nature of Vis--NIR measurement in water-dominated matrices, (ii) mismatch between effective receptive field (ERF) and the width of informative spectral structure, and (iii) validation design (including split strategy, hyperparameter tuning budget, and exposure to deployment-like shifts) acting as a hidden hyperparameter that can dominate model ranking. Building on evidence from published chemometrics and spectroscopy studies, we propose a conditional design framework that links architecture and preprocessing choices to spectral physics, dataset regime, and intended deployment scenario. Overall, the proposed perspective moves DL Chemometrics from template-driven architecture selection toward reproducible, physics-aware, and deployment-aligned model comparison.