iPINN for Broadband CARS Phase Retrieval: A Framework for Function Approximation and Inverse Modeling Problems in Nonlinear Spectroscopy
Abstract
Phase retrieval in broadband coherent anti-Stokes Raman spectroscopy (BCARS) is an ill-posed inverse problem. The Raman-like signal is encoded in the imaginary part of the resonant susceptibility, which mixes coherently with a non-resonant background (NRB) that varies across acquisitions. We introduce an inverse physics-informed neural network (iPINN) that predicts Lorentzian peak parameters from raw BCARS spectra and reconstructs the resonant susceptibility through a differentiable analytical forward model. A transformer encoder assigns spectral features to 24 learnable peak slots, and a multi-view consistency loss enforces invariance across NRB pattern, NRB strength, and noise. Unlike direct spectral regression approaches, the method retains accuracy under varying acquisition conditions. On a public benchmark, iPINN achieves the lowest error among the tested baselines (MAE 0.016 vs. next-best 0.046). On 28 zero-shot test spectra acquired across seven solvents and four focal positions, accuracy is depth-invariant in five of seven solvents. These results show that inverse parametric prediction with a differentiable physical decoder supports robust phase retrieval across measurement conditions.
Related
- Retrieval of the dephasing time in solids from spatial high-harmonic spectroscopy
- Entanglement-enhanced Ramsey spectroscopy of broadband signals
- Ultra-broadband quantum infrared spectroscopy extending to 20 {\textmu}m
- Local Broadband Gigahertz Spectroscopy via SQUID-on-Tip DC Readout
- Quantum environment afterglow from broadband excitation spectroscopy in superconducting qubits
BibTeX
@misc{vulchi2026ipinn,
title = {{iPINN for Broadband CARS Phase Retrieval: A Framework for Function Approximation and Inverse Modeling Problems in Nonlinear Spectroscopy}},
author = {Ravi Teja Vulchi and Carl Messerschmidt and Mohammadsadegh Vafaeinezhad and Rajendhar Junjuri and Tobias Meyer-Zedler and Juergen Popp and Thomas Bocklitz},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.00883},
eprint = {2609.00883},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.00883},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - iPINN for Broadband CARS Phase Retrieval: A Framework for Function Approximation and Inverse Modeling Problems in Nonlinear Spectroscopy
AU - Ravi Teja Vulchi
AU - Carl Messerschmidt
AU - Mohammadsadegh Vafaeinezhad
AU - Rajendhar Junjuri
AU - Tobias Meyer-Zedler
AU - Juergen Popp
AU - Thomas Bocklitz
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2609.00883
UR - https://arxiv.org/abs/2609.00883
AB - Phase retrieval in broadband coherent anti-Stokes Raman spectroscopy (BCARS) is an ill-posed inverse problem. The Raman-like signal is encoded in the imaginary part of the resonant susceptibility, which mixes coherently with a non-resonant background (NRB) that varies across acquisitions. We introduce an inverse physics-informed neural network (iPINN) that predicts Lorentzian peak parameters from raw BCARS spectra and reconstructs the resonant susceptibility through a differentiable analytical forward model. A transformer encoder assigns spectral features to 24 learnable peak slots, and a multi-view consistency loss enforces invariance across NRB pattern, NRB strength, and noise. Unlike direct spectral regression approaches, the method retains accuracy under varying acquisition conditions. On a public benchmark, iPINN achieves the lowest error among the tested baselines (MAE 0.016 vs. next-best 0.046). On 28 zero-shot test spectra acquired across seven solvents and four focal positions, accuracy is depth-invariant in five of seven solvents. These results show that inverse parametric prediction with a differentiable physical decoder supports robust phase retrieval across measurement conditions.
ER -