Track synthesis, materials and catalysis as it is published.
Search papers, labs, and topics across Lattice.
To test whether standard spectral fidelity metrics reliably reflect downstream model performance, this work introduces a controlled-perturbation framework measuring thirteen spectral quality metrics against three real-world downstream tasks (bacterial classification, sugar quantification, and mineral retrieval). The authors formulate two evaluation diagnostics—alignment gap, which tracks metric-to-task invariance across perturbation regimes, and ordering concordance, which tracks metric ranking accuracy of downstream performance degradation. The evaluation reveals stark, task-specific failure modes across common measures like MSE, Wasserstein distance, and peak F1, demonstrating that superior metric ordering concordance frequently fails to prevent severe downstream alignment drift under real-world spectral shifts.
Standard spectral quality metrics like MSE and Wasserstein distance often fail to track downstream model degradation, exposing a critical disconnect where high metric rankings do not guarantee better classification or regression performance.
Raman spectral preprocessing and enhancement are often evaluated by comparing output spectra with a reference. Interpreting these comparisons requires evidence that spectral quality measures reflect downstream task performance. We present a controlled-perturbation framework for testing this relationship. Five perturbation types (baseline distortion, independent noise, correlated noise, a global wavenumber shift, and nonlinear axis warping) generate paired changes in a spectral measure (metric harm) and in downstream performance (task harm). An alignment gap (AG) quantifies how much the relationship between metric harm and task harm changes with perturbation type. Ordering concordance (OC) measures how often a metric correctly ranks two conditions by their task harm. The framework evaluates thirteen outputs (MSE, RMSE, MAE, NMSE, spectral angle, Pearson correlation, Wasserstein distance, a structure-to-noise ratio, peak precision, recall, F1, artifact ratio, and missing ratio). Three public datasets provide bacterial classification, sugar-mixture quantification, and mineral identification tasks. PCA with logistic regression, partial least squares regression, and cosine library matching supply the task outcomes. Classifiers and calibrations are fitted either to unperturbed training spectra or to each perturbed training condition, then evaluated on the same perturbed test spectra. Mineral queries are compared with an unchanged or correspondingly perturbed library. The resulting comparisons identify task-specific strengths and limitations, including cases where better ordering does not accompany a smaller AG. Removing axis perturbations and comparing spectra on a common physical grid test how these findings depend on the evaluation design. The framework provides a reproducible procedure for assessing existing measures and testing new candidates against downstream task performance.