Wei Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wei Liu has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Liming Wu
Université Blaise Pascal
3 shared publications Fengwu Zhu
2 shared publications- Fengwu Zhu
School of Mathematics and Statistics Wuhan University 430072 Wuhan Hubei PR China
2 shared publications - Huijie Qiao
Southeast University
2 shared publications - Jianliang Zhai
School of Mathematical Sciences, University of Science and Technology of China, Hefei 230026, China
2 shared publications Jingyuan Li
2 shared publications
Latest Publications
- 21 August 2026
- Published by Elsevier in Communications in Nonlinear Science and Numerical Simulation
- Vol. 163
Solving McKean–Vlasov stochastic differential equations by a deep learning particle method
- 21 August 2026
- Published by Elsevier in Communications in Nonlinear Science and Numerical Simulation
- Vol. 163
Solving McKean–Vlasov stochastic differential equations by a deep learning particle method
- 19 August 2026
- Published by Elsevier in Statistics & Probability Letters
- Vol. 240
Cramér-type moderate deviations for a second-order continuous-time non-ergodic Gaussian autoregression
- 19 August 2026
- Published by Elsevier in Statistics & Probability Letters
- Vol. 240
Cramér-type moderate deviations for a second-order continuous-time non-ergodic Gaussian autoregression
- Open Access
Long term convergence rate of Smoluchowski-Kramers approximation by Stein's method
We consider the following second-order stochastic differential equation on $\mathbb{R}^{2d}$: \begin{equation*} dX_t^m=Y_t^mdt, \quad mdY_t^m=b(X_t^m)dt+σ(X_t^m)dB_t-Y^m_tdt, \end{equation*} where $X^m_t$ and $Y^m_t$ represent the position and velocity of a particle at time $t$, $m>0$ denotes its mass, $b:\mathbb{R}^d \rightarrow \mathbb{R}^d$ is the drift field, $σ:\mathbb{R}^d \rightarrow \mathbb{R}^{d \times d}$ is the diffusion coefficient, and $\{B_t\}_{t \ge 0}$ is a $d$-dimensional standard Brownian motion. The Smoluchowski--Kramers approximation states that as $m \rightarrow 0$, this system converges to the limiting equation: \begin{equation*} dX_t=b(X_t)dt+σ(X_t)dB_t. \end{equation*} Utilizing Stein's method, we prove that the $1$-Wasserstein distance between the invariant distribution of $X_t^m$ and that of its small-mass limit $X_t$ is of order $O(\sqrt{m}|\ln m|).$ Particularly, in the one-dimensional case, the convergence rate can be improved to $O(\sqrt{m}).$Open AccessLong term convergence rate of Smoluchowski-Kramers approximation by Stein's method
We consider the following second-order stochastic differential equation on $\mathbb{R}^{2d}$: \begin{equation*} dX_t^m=Y_t^mdt, \quad mdY_t^m=b(X_t^m)dt+σ(X_t^m)dB_t-Y^m_tdt, \end{equation*} where $X^m_t$ and $Y^m_t$ represent the position and velocity of a particle at time $t$, $m>0$ denotes its mass, $b:\mathbb{R}^d \rightarrow \mathbb{R}^d$ is the drift field, $σ:\mathbb{R}^d \rightarrow \mathbb{R}^{d \times d}$ is the diffusion coefficient, and $\{B_t\}_{t \ge 0}$ is a $d$-dimensional standard Brownian motion. The Smoluchowski--Kramers approximation states that as $m \rightarrow 0$, this system converges to the limiting equation: \begin{equation*} dX_t=b(X_t)dt+σ(X_t)dB_t. \end{equation*} Utilizing Stein's method, we prove that the $1$-Wasserstein distance between the invariant distribution of $X_t^m$ and that of its small-mass limit $X_t$ is of order $O(\sqrt{m}|\ln m|).$ Particularly, in the one-dimensional case, the convergence rate can be improved to $O(\sqrt{m}).$ - Open Access
Large and Moderate deviation principles for the Multivalued McKean-Vlasov SDEs with jumps
By using the weak convergence method, we establish the large and moderate deviation principles for the multivalued McKean-Vlasov SDEs with non-Lipschitz coefficients driven by Lévy noise in this paper. The Bihari's inequality is used to overcome the challenges arising from the non-Lipschitz conditions on the coefficients.
Open AccessLarge and Moderate deviation principles for the Multivalued McKean-Vlasov SDEs with jumps
By using the weak convergence method, we establish the large and moderate deviation principles for the multivalued McKean-Vlasov SDEs with non-Lipschitz coefficients driven by Lévy noise in this paper. The Bihari's inequality is used to overcome the challenges arising from the non-Lipschitz conditions on the coefficients.
- Open Access
General large deviations and functional iterated logarithm law for multivalued McKean-Vlasov stochastic differential equations
In this paper, we present sufficient conditions and criteria to establish general large and moderate deviation principles for multivalued McKean-Vlasov stochastic differential equations (SDEs in short) by means of the weak convergence approach, under non-Lipschit assumptions on the coefficents of the equations. Furthermore, by applying the large deviation estimates we obtain the functional iterated logarithm law for the solutions of multivalued McKean-Vlasov SDEs.
Open AccessGeneral large deviations and functional iterated logarithm law for multivalued McKean-Vlasov stochastic differential equations
In this paper, we present sufficient conditions and criteria to establish general large and moderate deviation principles for multivalued McKean-Vlasov stochastic differential equations (SDEs in short) by means of the weak convergence approach, under non-Lipschit assumptions on the coefficents of the equations. Furthermore, by applying the large deviation estimates we obtain the functional iterated logarithm law for the solutions of multivalued McKean-Vlasov SDEs.
- Open Access
Solving McKean-Vlasov Equation by deep learning particle method
We introduce a novel meshless simulation method for the McKean-Vlasov Stochastic Differential Equation (MV-SDE) utilizing deep learning, applicable to both self-interaction and interaction scenarios. Traditionally, numerical methods for this equation rely on the interacting particle method combined with techniques based on the Itô-Taylor expansion. The convergence rate of this approach is determined by two parameters: the number of particles $N$ and the time step size $h$ for each Euler iteration. However, for extended time horizons or equations with larger Lipschitz coefficients, this method is often limited, as it requires a significant increase in Euler iterations to achieve the desired precision $ε$. To overcome the challenges posed by the difficulty of parallelizing the simulation of continuous interacting particle systems, which involve solving high-dimensional coupled SDEs, we propose a meshless MV-SDE solver grounded in Physics-Informed Neural Networks (PINNs) that does not rely on the propagation of chaos result. Our method constructs a pseudo MV-SDE using Itô calculus, then quantifies the discrepancy between this equation and the original MV-SDE, with the error minimized through a loss function. This loss is controlled via an optimization algorithm, independent of the time step size, and we provide an error estimate for the loss function. The advantages of our approach are demonstrated through corresponding simulations.Open AccessSolving McKean-Vlasov Equation by deep learning particle method
We introduce a novel meshless simulation method for the McKean-Vlasov Stochastic Differential Equation (MV-SDE) utilizing deep learning, applicable to both self-interaction and interaction scenarios. Traditionally, numerical methods for this equation rely on the interacting particle method combined with techniques based on the Itô-Taylor expansion. The convergence rate of this approach is determined by two parameters: the number of particles $N$ and the time step size $h$ for each Euler iteration. However, for extended time horizons or equations with larger Lipschitz coefficients, this method is often limited, as it requires a significant increase in Euler iterations to achieve the desired precision $ε$. To overcome the challenges posed by the difficulty of parallelizing the simulation of continuous interacting particle systems, which involve solving high-dimensional coupled SDEs, we propose a meshless MV-SDE solver grounded in Physics-Informed Neural Networks (PINNs) that does not rely on the propagation of chaos result. Our method constructs a pseudo MV-SDE using Itô calculus, then quantifies the discrepancy between this equation and the original MV-SDE, with the error minimized through a loss function. This loss is controlled via an optimization algorithm, independent of the time step size, and we provide an error estimate for the loss function. The advantages of our approach are demonstrated through corresponding simulations. - 1 February 2024
- Published by World Scientific Pub Co Pte Ltd in Stochastics and Dynamics
- Vol. 24 (01)
Rate of convergence for the Smoluchowski–Kramers approximation for distribution-dependent SDEs driven by fractional Brownian motions
In this paper, we study the rate in the Smoluchowski–Kramers approximation for the solution of the following distribution-dependent SDE driven by fractional Brownian motion [ ] where [ ] denotes the law of [ ], [ ] is a [ ]-dimensional fractional Brownian motion with Hurst parameter [ ]. Based on the techniques of multiple integrals and Malliavin calculus, we provide an explicit bound on total variation distance for the rate of convergence.
- 1 February 2024
- Published by World Scientific Pub Co Pte Ltd in Stochastics and Dynamics
- Vol. 24 (01)
Rate of convergence for the Smoluchowski–Kramers approximation for distribution-dependent SDEs driven by fractional Brownian motions
In this paper, we study the rate in the Smoluchowski–Kramers approximation for the solution of the following distribution-dependent SDE driven by fractional Brownian motion [ ] where [ ] denotes the law of [ ], [ ] is a [ ]-dimensional fractional Brownian motion with Hurst parameter [ ]. Based on the techniques of multiple integrals and Malliavin calculus, we provide an explicit bound on total variation distance for the rate of convergence.
- 1 January 2024
- Published by American Institute of Mathematical Sciences (AIMS) in Discrete and Continuous Dynamical Systems-Series B
- Vol. 29 (5) , 2282-2297
Lipschitzian norms and functional inequalities for birth-death processes
In this paper, we consider a birth-death process with generator $ \mathcal{L} $ and reversible invariant probability measure $ \pi. $ We identify explicitly the Lipschitzian norm of the solution of the Poisson equation $ -\mathcal{L} G = g-\pi(g) $ for $ |g|\le\varphi $. This leads to some transportation-information inequalities, concentration inequalities and Cheeger-type isoperimetric inequalities. Several examples are provided to illustrate the results.- 1 January 2024
- Published by American Institute of Mathematical Sciences (AIMS) in Discrete and Continuous Dynamical Systems-Series B
- Vol. 29 (5) , 2282-2297
Lipschitzian norms and functional inequalities for birth-death processes
In this paper, we consider a birth-death process with generator $ \mathcal{L} $ and reversible invariant probability measure $ \pi. $ We identify explicitly the Lipschitzian norm of the solution of the Poisson equation $ -\mathcal{L} G = g-\pi(g) $ for $ |g|\le\varphi $. This leads to some transportation-information inequalities, concentration inequalities and Cheeger-type isoperimetric inequalities. Several examples are provided to illustrate the results.- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 54 (8) , 4539-4552
Reducing Mode Collapse With Monge–Kantorovich Optimal Transport for Generative Adversarial Networks
Mode collapse has been a persisting challenge in generative adversarial networks (GANs), and it directly affects the applications of GAN in many domains. Existing works that attempt to solve this problem have some serious limitations: models using optimal transport (OT) strategies (e.g., Wasserstein distance) lead to vanishing or exploding gradients; increasing the number of generators can cause several generators focusing on the same mode; and approaches that modify the loss also do not satisfactorily resolve mode collapse. In this article, we reduce mode collapse by formulating it as a Monge problem of OT map. We show that the Monge problem can be transformed to the distribution transformation problem in GAN, and a rectified affine neural network can be considered as a measurable function. In this way, we propose Monge GAN that uses this measurable function to transform the generated data distribution into the original data distribution. We utilize the Kantorovich formulation to obtain the OT cost, which is regarded as the OT distance between the two distributions. Finally, we conduct extensive experiments on both image and numerical datasets to validate our Monge GAN in reducing model collapse.
- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 54 (8) , 4539-4552
Reducing Mode Collapse With Monge–Kantorovich Optimal Transport for Generative Adversarial Networks
Mode collapse has been a persisting challenge in generative adversarial networks (GANs), and it directly affects the applications of GAN in many domains. Existing works that attempt to solve this problem have some serious limitations: models using optimal transport (OT) strategies (e.g., Wasserstein distance) lead to vanishing or exploding gradients; increasing the number of generators can cause several generators focusing on the same mode; and approaches that modify the loss also do not satisfactorily resolve mode collapse. In this article, we reduce mode collapse by formulating it as a Monge problem of OT map. We show that the Monge problem can be transformed to the distribution transformation problem in GAN, and a rectified affine neural network can be considered as a measurable function. In this way, we propose Monge GAN that uses this measurable function to transform the generated data distribution into the original data distribution. We utilize the Kantorovich formulation to obtain the OT cost, which is regarded as the OT distance between the two distributions. Finally, we conduct extensive experiments on both image and numerical datasets to validate our Monge GAN in reducing model collapse.