Adaptive weighting of Bayesian physics informed neural networks for multitask and multiscale forward and inverse problems
From MaRDI portal
(Redirected from Publication:6095075)
Abstract: In this paper, we present a novel methodology for automatic adaptive weighting of Bayesian Physics-Informed Neural Networks (BPINNs), and we demonstrate that this makes it possible to robustly address multi-objective and multi-scale problems. BPINNs are a popular framework for data assimilation, combining the constraints of Uncertainty Quantification (UQ) and Partial Differential Equation (PDE). The relative weights of the BPINN target distribution terms are directly related to the inherent uncertainty in the respective learning tasks. Yet, they are usually manually set a-priori, that can lead to pathological behavior, stability concerns, and to conflicts between tasks which are obstacles that have deterred the use of BPINNs for inverse problems with multi-scale dynamics. The present weighting strategy automatically tunes the weights by considering the multi-task nature of target posterior distribution. We show that this remedies the failure modes of BPINNs and provides efficient exploration of the optimal Pareto front. This leads to better convergence and stability of BPINN training while reducing sampling bias. The determined weights moreover carry information about task uncertainties, reflecting noise levels in the data and adequacy of the PDE model. We demonstrate this in numerical experiments in Sobolev training, and compare them to analytically -optimal baseline, and in a multi-scale Lokta-Volterra inverse problem. We eventually apply this framework to an inpainting task and an inverse problem, involving latent field recovery for incompressible flow in complex geometries.
Recommendations
- Surrogate modeling for Bayesian inverse problems based on physics-informed neural networks
- Efficient Bayesian physics informed neural networks for inverse problems via ensemble Kalman inversion
- Auto-weighted Bayesian physics-informed neural networks and robust estimations for multitask inverse problems in pore-scale imaging of dissolution
- Physics and equality constrained artificial neural networks: application to forward and inverse problems with multi-fidelity data fusion
- B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
Cites work
- A velocity-vorticity method for highly viscous 3D flows with application to digital rock physics
- Adaptive multi-fidelity polynomial chaos approach to Bayesian inference in inverse problems
- Adversarial uncertainty quantification in physics-informed neural networks
- An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems
- B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
- Bayesian physics informed neural networks for real-world nonlinear dynamical systems
- Contribution to the Discussion of the Paper ‘Geodesic Monte Carlo on Embedded Manifolds’
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
- Learning partial differential equations via data discovery and sparse optimization
- Model selection in Bayesian neural networks via horseshoe priors
- Multi-fidelity Bayesian neural networks: algorithms and applications
- Multi-variance replica exchange SGMCMC for inverse and forward problems via Bayesian PINN
- On the geometric ergodicity of Hamiltonian Monte Carlo
- Optimally weighted loss functions for solving PDEs with neural networks
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Physics-informed semantic inpainting: application to geostatistical modeling
- Riemann manifold Langevin and Hamiltonian Monte Carlo methods. With discussion and authors' reply
- Sobolev training of thermodynamic-informed neural networks for interpretable elasto-plasticity models with level set hardening
- Spatially distributed control for optimal drag reduction of the flow past a circular cylinder
- Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
- The geometric foundations of Hamiltonian Monte Carlo
- The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo
- Uncertainty quantification in scientific machine learning: methods, metrics, and comparisons
- Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
- Using DC PSE operator discretization in Eulerian meshless collocation methods improves their robustness in complex geometries
- When and why PINNs fail to train: a neural tangent kernel perspective
Cited in
(8)- Variational inference for nonlinear inverse problems via neural net kernels: comparison to Bayesian neural networks, application to topology optimization
- Multi-variance replica exchange SGMCMC for inverse and forward problems via Bayesian PINN
- Surrogate modeling for Bayesian inverse problems based on physics-informed neural networks
- Auto-weighted Bayesian physics-informed neural networks and robust estimations for multitask inverse problems in pore-scale imaging of dissolution
- Adaptive residual splitting in PINNs for solving complex PDEs
- Physics-informed non-intrusive reduced-order modeling of parameterized dynamical systems
- Invariant deep neural networks under the finite group for solving partial differential equations
- Advancements and future directions in loss function designs for physics-informed neural networks: a comprehensive review
This page was built for publication: Adaptive weighting of Bayesian physics informed neural networks for multitask and multiscale forward and inverse problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6095075)
