Seok-Bae Yun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Seok-Bae Yun has a total of 55 co-authors. Here are the top co-authors by number of shared publications:
- Seung Yeon Cho
Department of Mathematics, Gyeongsang National University, Jinju, Republic of Korea
13 shared publications - Byung-Hoon Hwang
Sangmyung Univ, Dept Math Educ, 20 Hongjimun 2 Gil, Seoul 03016, South Korea
8 shared publications - Doheon Kim
Department of Mathematical Data Science, Hanyang University, Ansan, Republic of Korea
7 shared publications - Jin Woo Jang
Department of Mathematics, Pohang University of Science and Technology (POSTECH), Pohang, South Korea
7 shared publications - Hyeong-Ohk Bae
Department of Financial Engineering, Ajou University, Suwon-si, South Korea
6 shared publications - Giovanni Russo
Department of Mathematics and Computer Science, University of Catania, Catania, Italy
5 shared publications
Latest Publications
- Open Access
- 19 October 2026
- Published by Elsevier in SSRN Electronic Journal
The Boltzmann-BGK model with nontrivial collision frequency near vacuum
We study the Cauchy problem for the Boltzmann-BGK model in the three-dimensional whole space with collision frequency $\nu(f)=\rho^\alpha T^\beta$, where $\alpha\in(1/3,1]$, $\beta\in[0,1]$, and $3\alpha+2\beta\geq2$. For nonnegative initial data with finite mass and energy, sufficiently small in suitable polynomial-weighted $L^\infty$ norms, we establish global existence and uniqueness of mild solutions near vacuum without uniform positive lower bounds on density or temperature. The proof combines two mechanisms. First, a phase-space weight invariant along free transport yields time-integrable dispersive decay for the collision frequency and gain term, controlling nonlinear growth and closing global weighted estimates. Second, the collision frequency compensates for the singular dependence of the local Maxwellian on macroscopic fields, yielding a weighted $L^1$ Lipschitz estimate for the relaxation operator with a constant depending only on weighted upper bounds. Together, these estimates provide global well-posedness and uniform weighted bounds.- 19 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Boltzmann-BGK model with nontrivial collision frequency near vacuum
We study the Cauchy problem for the Boltzmann-BGK model in the three-dimensional whole space with collision frequency $\nu(f)=\rho^\alpha T^\beta$, where $\alpha\in(1/3,1]$, $\beta\in[0,1]$, and $3\alpha+2\beta\geq2$. For nonnegative initial data with finite mass and energy, sufficiently small in suitable polynomial-weighted $L^\infty$ norms, we establish global existence and uniqueness of mild solutions near vacuum without uniform positive lower bounds on density or temperature. The proof combines two mechanisms. First, a phase-space weight invariant along free transport yields time-integrable dispersive decay for the collision frequency and gain term, controlling nonlinear growth and closing global weighted estimates. Second, the collision frequency compensates for the singular dependence of the local Maxwellian on macroscopic fields, yielding a weighted $L^1$ Lipschitz estimate for the relaxation operator with a constant depending only on weighted upper bounds. Together, these estimates provide global well-posedness and uniform weighted bounds. - Open Access
- 1 September 2026
- Published by Elsevier in SSRN Electronic Journal
Convergence Analysis of Unsupervised Legendre-Galerkin Operator Networks for Solving Parametric PDEs
In this paper, we present the convergence analysis of a novel operator learning approach called Unsupervised Legendre-Galerkin Neural Networks (ULGNet), a deep learning-based numerical method for solving partial differential equations (PDEs). Unlike existing deep learning-based numerical methods for solving PDEs, ULGNet expresses the solution as a spectral expansion with respect to the Legendre basis and predicts the coefficients with deep neural networks by solving a variational residual minimization problem. Based on the observation that the corresponding loss function is equivalent to the residual induced by the linear algebraic system, depending on the choice of basis functions, we prove that the minimizer of the discrete loss function converges to the weak solution of the PDEs. Numerical evidence will also be provided to support the theoretical result. Key technical tools include a variant of the universal approximation theorem for bounded neural networks, the analysis of stiffness and mass matrices, and the application of the uniform law of large numbers in terms of the Rademacher complexity.
- 1 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessConvergence Analysis of Unsupervised Legendre-Galerkin Operator Networks for Solving Parametric PDEs
In this paper, we present the convergence analysis of a novel operator learning approach called Unsupervised Legendre-Galerkin Neural Networks (ULGNet), a deep learning-based numerical method for solving partial differential equations (PDEs). Unlike existing deep learning-based numerical methods for solving PDEs, ULGNet expresses the solution as a spectral expansion with respect to the Legendre basis and predicts the coefficients with deep neural networks by solving a variational residual minimization problem. Based on the observation that the corresponding loss function is equivalent to the residual induced by the linear algebraic system, depending on the choice of basis functions, we prove that the minimizer of the discrete loss function converges to the weak solution of the PDEs. Numerical evidence will also be provided to support the theoretical result. Key technical tools include a variant of the universal approximation theorem for bounded neural networks, the analysis of stiffness and mass matrices, and the application of the uniform law of large numbers in terms of the Rademacher complexity.
- 28 April 2026
- Published by Springer Nature in Journal of Statistical Physics
- Vol. 193 (5)
Stationary Boltzmann Equation for Polyatomic Gases in a slab
- 28 April 2026
- Published by Springer Nature in Journal of Statistical Physics
- Vol. 193 (5)
Stationary Boltzmann Equation for Polyatomic Gases in a slab
- 8 April 2026
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Mathematical Analysis
- Vol. 58 (2) , 1530-1570
Large Amplitude Problem of BGK Model: Relaxation to Quadratic Nonlinearity
The Bhatnagar–Gross–Krook (BGK) equation is a relaxation model of the Boltzmann equation which is widely used in place of the Boltzmann equation for the simulation of various kinetic flow problems. In this work, we study the asymptotic stability of the BGK model when the initial data is not necessarily close to the global equilibrium pointwisely. Due to the highly nonlinear structure of the relaxation operator, the argument developed to derive the bootstrap estimate for the Boltzmann equation leads to a weaker estimate in the case of the BGK model, which does not exclude the possible blow-up of the perturbation. To overcome this issue, we carry out a refined analysis of the macroscopic fields to guarantee that the system transits from a highly nonlinear regime into a quadratic nonlinear regime after a long but finite time, in which the highly nonlinear perturbative term relaxes to essentially quadratic nonlinearity.
- 8 April 2026
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Mathematical Analysis
- Vol. 58 (2) , 1530-1570
Large Amplitude Problem of BGK Model: Relaxation to Quadratic Nonlinearity
The Bhatnagar–Gross–Krook (BGK) equation is a relaxation model of the Boltzmann equation which is widely used in place of the Boltzmann equation for the simulation of various kinetic flow problems. In this work, we study the asymptotic stability of the BGK model when the initial data is not necessarily close to the global equilibrium pointwisely. Due to the highly nonlinear structure of the relaxation operator, the argument developed to derive the bootstrap estimate for the Boltzmann equation leads to a weaker estimate in the case of the BGK model, which does not exclude the possible blow-up of the perturbation. To overcome this issue, we carry out a refined analysis of the macroscopic fields to guarantee that the system transits from a highly nonlinear regime into a quadratic nonlinear regime after a long but finite time, in which the highly nonlinear perturbative term relaxes to essentially quadratic nonlinearity.
- Open Access
- 17 March 2026
- Published by Elsevier in SSRN Electronic Journal
Steady Solutions to the Relativistic Boltzmann Equation in a Slab
We study steady solutions to the relativistic Boltzmann equation with hard-sphere interactions in a slab geometry. For non-negative inflow boundary conditions, we prove the existence and uniqueness of a stationary solution in a weighted $L^1_p L^\infty_{x_1}$ framework, together with exponential decay in momentum. Our analysis treats the full slab domain and does not rely on any smallness assumption on the width. We establish sharp coercivity and continuity estimates for the collision frequency, together with weighted convolution and pointwise bounds for the nonlinear gain term. These estimates generate and propagate a $(-\Delta_p)^{-1}$-type regularity within the solution framework, which plays a crucial role in the existence and uniqueness argument. In addition, we obtain uniform weighted integrability of the solution over arbitrary two-dimensional hyperplanes through the origin. This hyperplane estimate is derived as a genuinely \textit{a posteriori} regularity property, without imposing any \textit{a priori} hyperplane bounds, and follows from a Lorentz-invariant geometric reduction.- 17 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSteady Solutions to the Relativistic Boltzmann Equation in a Slab
We study steady solutions to the relativistic Boltzmann equation with hard-sphere interactions in a slab geometry. For non-negative inflow boundary conditions, we prove the existence and uniqueness of a stationary solution in a weighted $L^1_p L^\infty_{x_1}$ framework, together with exponential decay in momentum. Our analysis treats the full slab domain and does not rely on any smallness assumption on the width. We establish sharp coercivity and continuity estimates for the collision frequency, together with weighted convolution and pointwise bounds for the nonlinear gain term. These estimates generate and propagate a $(-\Delta_p)^{-1}$-type regularity within the solution framework, which plays a crucial role in the existence and uniqueness argument. In addition, we obtain uniform weighted integrability of the solution over arbitrary two-dimensional hyperplanes through the origin. This hyperplane estimate is derived as a genuinely \textit{a posteriori} regularity property, without imposing any \textit{a priori} hyperplane bounds, and follows from a Lorentz-invariant geometric reduction. - 5 November 2025
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Mathematical Analysis
- Vol. 57 (6) , 6361-6402
Tsallis-BGK Model Near a Global Equilibrium
We study global-in-time well-posedness and asymptotic behavior of a classical solution to the generalized BGK model, which will be referred as the Tsallis–BGK model, whose relaxation distribution is the Tsallis [Formula: see text]-Maxwellian distribution. The Tsallis distribution is a generalization of the classical Maxwellian distribution toward heavy-tailed and to light-tailed distributions depending on the choice of the nonextensivity index [Formula: see text]. To establish well-posedness of the Tsallis–BGK model, we linearize it near the global [Formula: see text]-Maxwellian and use a nonlinear energy estimate to obtain local well-posedness, which can be extended to the global-in-time by using the coercivity of the linearization operator. The asymptotic behavior follows from the energy estimate. We also verify numerically that such linearization argument breaks down when [Formula: see text] due to the discrepancy of the velocity support.
- 5 November 2025
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Mathematical Analysis
- Vol. 57 (6) , 6361-6402
Tsallis-BGK Model Near a Global Equilibrium
We study global-in-time well-posedness and asymptotic behavior of a classical solution to the generalized BGK model, which will be referred as the Tsallis–BGK model, whose relaxation distribution is the Tsallis [Formula: see text]-Maxwellian distribution. The Tsallis distribution is a generalization of the classical Maxwellian distribution toward heavy-tailed and to light-tailed distributions depending on the choice of the nonextensivity index [Formula: see text]. To establish well-posedness of the Tsallis–BGK model, we linearize it near the global [Formula: see text]-Maxwellian and use a nonlinear energy estimate to obtain local well-posedness, which can be extended to the global-in-time by using the coercivity of the linearization operator. The asymptotic behavior follows from the energy estimate. We also verify numerically that such linearization argument breaks down when [Formula: see text] due to the discrepancy of the velocity support.
- Open Access
Relativistic BGK model for reactive gas mixtures
We propose a BGK-type kinetic model for relativistic reactive gas mixtures. This model serves as a computationally tractable yet physically consistent alternative to the corresponding Boltzmann equation. The relaxation operator is constructed to ensure that the model correctly satisfies the conservation laws and relaxes to the proper equilibrium: a Jüttner distribution characterized by a common temperature, velocity, and chemical potentials that obey the law of mass action. Furthermore, we prove that the model satisfies an H-theorem with the same entropy functional as the original Boltzmann equation. Finally, numerical simulations are presented, which confirm that the model preserves the conserved quantities and exhibits entropy decay towards the proper Jüttner equilibrium.
Open AccessRelativistic BGK model for reactive gas mixtures
We propose a BGK-type kinetic model for relativistic reactive gas mixtures. This model serves as a computationally tractable yet physically consistent alternative to the corresponding Boltzmann equation. The relaxation operator is constructed to ensure that the model correctly satisfies the conservation laws and relaxes to the proper equilibrium: a Jüttner distribution characterized by a common temperature, velocity, and chemical potentials that obey the law of mass action. Furthermore, we prove that the model satisfies an H-theorem with the same entropy functional as the original Boltzmann equation. Finally, numerical simulations are presented, which confirm that the model preserves the conserved quantities and exhibits entropy decay towards the proper Jüttner equilibrium.
- 29 August 2025
- Published by Springer Nature in Journal of Scientific Computing
- Vol. 105 (1) , 1-25
A conservative semi-Lagrangian scheme for the ellipsoidal BGK model of the Boltzmann equation
- 29 August 2025
- Published by Springer Nature in Journal of Scientific Computing
- Vol. 105 (1) , 1-25
A conservative semi-Lagrangian scheme for the ellipsoidal BGK model of the Boltzmann equation
- 22 May 2025
- Published by Springer Nature in Journal of Global Optimization
- Vol. 92 (3) , 713-736
On the convergence result of the gradient-push algorithm on directed graphs with constant stepsize
- 22 May 2025
- Published by Springer Nature in Journal of Global Optimization
- Vol. 92 (3) , 713-736
On the convergence result of the gradient-push algorithm on directed graphs with constant stepsize
- 4 April 2025
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Scientific Computing
- Vol. 47 (2) , C451-C474
Separable Physics-Informed Neural Networks for Solving the BGK Model of the Boltzmann Equation
In this study, we present a new approach based on separable physics-informed neural networks (SPINNs), specifically designed to efficiently solve the Bhatnagar–Gross–Krook (BGK) model. While the mesh-free nature of physics-informed neural networks (PINNs) offers significant advantages for handling high-dimensional partial differential equations, applying quadrature rules for accurate integral evaluation in the BGK operator can compromise these benefits and increase computational costs. To address this issue, we leverage the canonical polyadic decomposition structure of SPINNs and the linear nature of moment calculation to significantly reduce the computational expense associated with quadrature rules. However, the multiscale nature of the particle density function poses challenges for precisely approximating macroscopic moments using neural networks. To overcome this, we introduce SPINN-BGK, a specialized variant of SPINNs that fuses SPINNs with Gaussian functions and utilizes a relative loss approach. This modification enables SPINNs to decay as rapidly as Maxwellian distributions, enhancing the accuracy of macroscopic moment approximations. The relative loss design ensures that both large- and small-scale features are effectively captured by the SPINNs. The effectiveness of our approach is validated through six numerical experiments, including a complex three-dimensional Riemann problem. These experiments demonstrate the potential of our method to efficiently and accurately tackle intricate challenges in computational physics, offering significant improvements in computational efficiency and solution accuracy.
- 4 April 2025
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Scientific Computing
- Vol. 47 (2) , C451-C474
Separable Physics-Informed Neural Networks for Solving the BGK Model of the Boltzmann Equation
In this study, we present a new approach based on separable physics-informed neural networks (SPINNs), specifically designed to efficiently solve the Bhatnagar–Gross–Krook (BGK) model. While the mesh-free nature of physics-informed neural networks (PINNs) offers significant advantages for handling high-dimensional partial differential equations, applying quadrature rules for accurate integral evaluation in the BGK operator can compromise these benefits and increase computational costs. To address this issue, we leverage the canonical polyadic decomposition structure of SPINNs and the linear nature of moment calculation to significantly reduce the computational expense associated with quadrature rules. However, the multiscale nature of the particle density function poses challenges for precisely approximating macroscopic moments using neural networks. To overcome this, we introduce SPINN-BGK, a specialized variant of SPINNs that fuses SPINNs with Gaussian functions and utilizes a relative loss approach. This modification enables SPINNs to decay as rapidly as Maxwellian distributions, enhancing the accuracy of macroscopic moment approximations. The relative loss design ensures that both large- and small-scale features are effectively captured by the SPINNs. The effectiveness of our approach is validated through six numerical experiments, including a complex three-dimensional Riemann problem. These experiments demonstrate the potential of our method to efficiently and accurately tackle intricate challenges in computational physics, offering significant improvements in computational efficiency and solution accuracy.