Discovery of unobservable parameters via physical embedding

TL;DR

PEIL is a physics-embedded inverse learning method that estimates unobservable parameters from indirect measurements, enabling improved signal reconstruction and zero-shot generalization across physical domains.

Contribution

Introduces PEIL, a novel approach that leverages physics-based inverse operators and learned estimators to recover unobservable parameters without direct supervision.

Findings

PEIL outperforms baselines in wireless communication with over 20-fold data reduction.

PEIL discovers interpretable coil sensitivity maps in MRI without calibration scans.

PEIL achieves zero-shot generalization across different physical systems.

Abstract

Recovering a source signal from indirect measurements often requires estimating latent parameters, such as wireless channel states or MRI coil sensitivities, that cannot be directly observed. Here, we introduce Physics-Embedded Inverse Learning (PEIL), in which a learned estimator predicts these parameters and a fixed, physics-based inverse operator uses them to reconstruct the signal, so that training requires only the source signal as supervision. In systems where multiple parameter combinations can reconstruct the signal equally well, the estimator exploits this freedom to coordinate parameters that compensate for residual modelling errors rather than match ground-truth parameters. In high-mobility wireless communications, PEIL discovers task-optimal configurations that outperform baselines given access to ground-truth parameters, enabling zero-shot generalisation and over 20-fold reduction in training data relative to supervised baselines. To test whether these properties extend across physical domains, we apply PEIL to parallel MRI, where it discovers physically interpretable coil sensitivity maps without calibration scans, yielding reconstructions grounded purely in acquired measurements. These results demonstrate that non-identifiability, conventionally a liability, becomes a resource when the learning objective targets reconstruction quality rather than parameter accuracy.