Skip to main content

# Physics-Informed Machine Learning

2 related articles

Subscribe to updates

arXiv Nov 5, 2026
Influence scores for traffic forecasts must pass through conservation constraints: classic rankings shift, and removing the top ten stations raises constrained-model MAE by 44.2%
When a traffic forecasting model enforces junction flow conservation at its output, a training sample's influence on predictions no longer travels only through the loss but also through the constraint projection. Classical influence functions assume unconstrained learning and cannot separate perturbations the physics keeps from those the projection removes. This work computes influence inside the feasible set, ranks station sensitivity on hourly PeMS District 3 data covering 318 freeway links, and shows that withholding the ten highest-ranked stations raises the constrained model's MAE by 44.2%.
When a traffic forecasting model enforces junction flow conservation at its output, a training sample's influence on predictions no longer travels only through the loss but also through the constraint projection. Classical influence functions assume unconstrained learning and cannot separate perturbations the physics keeps from those the projection removes. This work computes influence inside the feasible set, ranks station sensitivity on hourly PeMS District 3 data covering 318 freeway links, and shows that withholding the ten highest-ranked stations raises the constrained model's MAE by 44.2%.
University of Southern California University of California, Davis
Hamidreza Attari（First author）
Influence Functions Physics-Informed Machine Learning HardNet-Aff HardNetAff-STGNN Traffic Flow Forecasting
发表出处待核验 Sep 26, 2026
AI across the nuclear lifecycle: moving from isolated demonstrations to trustworthy decision support requires assessing complete workflows, not just predictive models
This review surveys evidence on artificial intelligence across the nuclear engineering lifecycle—covering nuclear datasets and computational infrastructure, surrogate and physics-informed modeling, digital twins, monitoring and prognostics, optimization and control, and trustworthy AI—reporting that learning-based methods can accelerate high-fidelity calculations, extract information from multivariate measurements, and support decisions in reactor operation, maintenance, waste management, and environmental assessment, while exposing recurring limitations such as scarce abnormal-condition data, differences between simulated and physical systems, uncertain generalization, and incomplete evaluation of downstream decisions, and arguing that progress depends on assessing complete AI-enabled wor
This review surveys evidence on artificial intelligence across the nuclear engineering lifecycle—covering nuclear datasets and computational infrastructure, surrogate and physics-informed modeling, digital twins, monitoring and prognostics, optimization and control, and trustworthy AI—reporting that learning-based methods can accelerate high-fidelity calculations, extract information from multivariate measurements, and support decisions in reactor operation, maintenance, waste management, and environmental assessment, while exposing recurring limitations such as scarce abnormal-condition data, differences between simulated and physical systems, uncertain generalization, and incomplete evaluation of downstream decisions, and arguing that progress depends on assessing complete AI-enabled wor
University of Michigan
M. I. Radaideh（First author, Corresponding author）
AI in Nuclear Engineering Physics-Informed Machine Learning Digital Twin

You’re all caught up