Leveraging Multi-time Hamilton-Jacobi PDEs for Certain Scientific Machine Learning Problems
From MaRDI portal
Abstract: Hamilton-Jacobi partial differential equations (HJ PDEs) have deep connections with a wide range of fields, including optimal control, differential games, and imaging sciences. By considering the time variable to be a higher dimensional quantity, HJ PDEs can be extended to the multi-time case. In this paper, we establish a novel theoretical connection between specific optimization problems arising in machine learning and the multi-time Hopf formula, which corresponds to a representation of the solution to certain multi-time HJ PDEs. Through this connection, we increase the interpretability of the training process of certain machine learning applications by showing that when we solve these learning problems, we also solve a multi-time HJ PDE and, by extension, its corresponding optimal control problem. As a first exploration of this connection, we develop the relation between the regularized linear regression problem and the Linear Quadratic Regulator (LQR). We then leverage our theoretical connection to adapt standard LQR solvers (namely, those based on the Riccati ordinary differential equations) to design new training approaches for machine learning. Finally, we provide some numerical examples that demonstrate the versatility and possible computational advantages of our Riccati-based approach in the context of continual learning, post-training calibration, transfer learning, and sparse dynamics identification.
Recommendations
- Leveraging viscous Hamilton-Jacobi PDEs for uncertainty quantification in scientific machine learning
- Overcoming the curse of dimensionality for some Hamilton-Jacobi partial differential equations via neural network architectures
- On decomposition models in imaging sciences and multi-time Hamilton-Jacobi partial differential equations
- A functional approach to interpreting the role of the adjoint equation in machine learning
- A mean-field optimal control formulation of deep learning
Cited in
(7)- A functional approach to interpreting the role of the adjoint equation in machine learning
- A comprehensive and FAIR comparison between MLP and KAN representations for differential equations and operator networks
- Leveraging viscous Hamilton-Jacobi PDEs for uncertainty quantification in scientific machine learning
- Uncertainty quantification for noisy inputs-outputs in physics-informed neural networks and neural operators
- NeuroSEM: a hybrid framework for simulating multiphysics problems by coupling PINNs and spectral elements
- A supervised learning scheme for computing Hamilton-Jacobi equation via density coupling
- Fast meta-solvers for 3D complex-shape scatterers using neural operators trained on a non-scattering problem
This page was built for publication: Leveraging Multi-time Hamilton-Jacobi PDEs for Certain Scientific Machine Learning Problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6430531)
