Wrong Operator or Blind Design? A Post-Training Diagnostic for Physics-Informed Coefficient Learning
Abstract
Physics-informed and hybrid models integrate data and physics, and read learned coefficients as physical. Yet a small prediction error guarantees neither that the postulated operator is right, nor that the design identifies the coefficient, nor that training reached its minimum. On one parabolic benchmark where a sensor aggregates several diffusion rates and the truth is known, a correct operator recovers the coefficient to machine precision. A wrong one leaves it 30% off while the in-domain error is 1.36 times the noise, and drifts 27% across windows. At 7% noise the error falls below the noise; the coefficient stays 30% off. Without the truth, a post-training coefficient diagnostic questions the trained model with three classical statistics, without oracle; the first two need no refit. Does it explain the data? Can the design resolve the named physics? Does the coefficient depend on the window? The specification test keeps its pre-registered false-alarm ceiling and rejects every replicate of both wrong operators tested; a network frozen at its minimum returns the same verdicts. Where the design is blind, a wrong operator confined to the unseen rate leaves the test mute; the design reading flags it beforehand. A network stopped short of its minimum mimics a wrong operator; an analytic second fit separates the two. The test has a notch against a missing reaction term, miscalibrates under correlated noise and loses power beyond 7% noise. Each warning names its remedy (operator, experiment or training); silence certifies nothing.
Related
- Preconditioned Physics-Informed Neural Operator Training
- Finance-Informed Operator Learning for Option Pricing with Quantum-Compatible Realizations
- Physics-informed random neural architectures on compact boundaryless multiply connected manifolds for random-operator models in machine learning and uncertainty quantification
- Physics-informed random neural architectures on compact boundaryless multiply connected manifolds for random-operator models in machine learning and uncertainty quantification
- A partial data coefficient identification inverse problem for a semilinear damped wave operator
BibTeX
@misc{fock2026wrong,
title = {{Wrong Operator or Blind Design? A Post-Training Diagnostic for Physics-Informed Coefficient Learning}},
author = {Eric Fock},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2608.16925},
eprint = {2608.16925},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.16925},
}
RIS
TY - UNPB
TI - Wrong Operator or Blind Design? A Post-Training Diagnostic for Physics-Informed Coefficient Learning
AU - Eric Fock
PY - 2026
DA - 2026/10/14
PB - arXiv
DO - 10.48550/arxiv.2608.16925
UR - https://arxiv.org/abs/2608.16925
AB - Physics-informed and hybrid models integrate data and physics, and read learned coefficients as physical. Yet a small prediction error guarantees neither that the postulated operator is right, nor that the design identifies the coefficient, nor that training reached its minimum. On one parabolic benchmark where a sensor aggregates several diffusion rates and the truth is known, a correct operator recovers the coefficient to machine precision. A wrong one leaves it 30% off while the in-domain error is 1.36 times the noise, and drifts 27% across windows. At 7% noise the error falls below the noise; the coefficient stays 30% off. Without the truth, a post-training coefficient diagnostic questions the trained model with three classical statistics, without oracle; the first two need no refit. Does it explain the data? Can the design resolve the named physics? Does the coefficient depend on the window? The specification test keeps its pre-registered false-alarm ceiling and rejects every replicate of both wrong operators tested; a network frozen at its minimum returns the same verdicts. Where the design is blind, a wrong operator confined to the unseen rate leaves the test mute; the design reading flags it beforehand. A network stopped short of its minimum mimics a wrong operator; an analytic second fit separates the two. The test has a notch against a missing reaction term, miscalibrates under correlated noise and loses power beyond 7% noise. Each warning names its remedy (operator, experiment or training); silence certifies nothing.
ER -