The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational Problems
1,879 citations recorded by the source corpus.
We propose a deep learning-based method, the Deep Ritz Method, for numerically solving variational problems, particularly the ones that arise from partial differential equations. The Deep Ritz Method is naturally nonlinear, naturally adaptive and has the potential to work in rather high dimensions. The framework is quite simple and fits well with the stochastic gradient descent method used in deep learning. We illustrate the method on several problems including some eigenvalue problems.
Subfields: Artificial Intelligence, Mathematical Physics, Statistical and Nonlinear Physics
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- G. Em Karniadakis
- Paris Perdikaris
- Anima Anandkumar
- Lu Lu
- Zong-Yi Li
- Kamyar Azizzadenesheli
- Yang Liu
- Xuhui Meng
In Mathematical Physics
- Zhiping Mao
- Nikola B. Kovachki
- Gianluigi Rozza
- Sharmila Karumuri
- Yaohua Zang
- Somdatta Goswami
- Nicholas Geneva
- Chunyue Zheng
In Statistical and Nonlinear Physics
Citations in the corpus
- Physics informed deep learning for computational elastodynamics without labeled data (2020, methodology): “ize the residual loss function of the PDE by the residuals of initial and boundary values via Lagrange multipliers in a \soft" manner. This strategy has been widely used in existing PINN methods [8,9,19,37]. In this case, the physics loss function J pcan be constructed with three components (i.”
- Weak Adversarial Networks for High-dimensional Partial Differential Equations (2019, methodology): “s of nonlinear PDEs by forwardbackward stochastic dierential equations is proposed and studied In [5, 6, 7, 8]. Another appealing approach that exploits the variational form of PDEs is considered in [9, 10, 11]. In [9], a committer function is parameterized by a neural network whose weights are obta”
- Neural Operator: Graph Kernel Network for Partial Differential Equations (2020, background): “…need only be trained once on the equations set { a j , u j } Nj =1 ; then, obtaining a solution for a new a ∼ µ , only requires a forward pass of the network, alleviating the major computational issues incurred in [E and Yu, 2018, Raissi et al., 2019, Herrmann et al., 2020, Bar and Sochen, 2019].”
- A Deep Collocation Method for the Bending Analysis of Kirchhoff Plate (2021, methodology): “Also, E and Yu proposed a Deep Ritz method for solving variational problems arising from partial differential equations [30].”
- Transolver: A Fast Transformer Solver for PDEs on General Geometries (2024, methodology): “Physics-informed neural networks This paradigm formalizes PDE constraints, including equations, initial and boundary conditions as objective functions of deep models (Weinan & Yu, 2017; Raissi et al., 2019; Wang et al., 2020a;b).”
- Neural operators for accelerating scientific simulations and design (2023, neutral): “Implicit Neural Representation (PINN and NeRF): The concept of using neural networks to represent continuous functions, known as implicit neural networks, has a rich history [53].”
- Transfer learning enhanced physics informed neural network for phase-field modeling of fracture (2019, background): “One major advantage of the proposed variational energy formulation resides in the fact that it requires derivatives one order lower than in the conventional residual minimization approach [9].”
- Exact imposition of boundary conditions with distance functions in physics-informed deep neural networks (2021, methodology): “In PINN-based deep collocation [6] and deep Ritz [9] methods, the inexact imposition of boundary conditions adversely affects the training of the neural network as well as the accuracy of the method [16–18].”
- MIONet: Learning multiple-input operators via tensor product (2022, methodology): “As an active area of research in SciML, different methods have been developed to solve ordinary and partial differential equations (ODEs and PDEs) by parameterizing the solutions via neural networks (NNs), such as physics-informed NNs (PINNs) [48, 30, 35, 37, 47], deep Ritz method [43], and deep Gal”
- Fourier Neural Operator for Parametric Partial Differential Equations (2020, background): “The second approach directly parameterizes the solution function as a neural network (E & Yu, 2018; Raissi et al., 2019; Bar & Sochen, 2019; Smith et al., 2020; Pan & Duraisamy, 2020).”