This is the current schedule for the Nonlinear PDE seminar in Texas A&M University.
Please see the event description in the embedded calendar for ZOOM link access etc. You are more than welcome to subscribe the calendar by clicking the + button on the bottom right of the calendar, or go to this link to open the calendar in a new page.
All times showed below are local Texas time (CDT/CST). Please consult WorldTimeBuddy for your local time.
Please contact Minh-Binh Tran at minhbinh_at_tamu_dot_edu or Xin Liu at xliu23_at_tamu_dot_edu for any question.
Title: The Fuzzy Landau equation and the Fisher information.
Abstract: The Landau equation was introduced by Lev Landau in 1936 as a modification of the Boltzmann equation to specific applications in plasma physics, and describes the interactions and collisions among charged particles in a plasma. The mathematical investigation of the Landau equation has been active for several decades, with researchers exploring various aspects of its behavior and properties. While the homogeneous version of this equation has been fully understood, the inhomogeneous equation remains a very difficult problem. In this talk I will present the first global well-posedness theory for a fuzzy version of the inhomogeneous Landau equation. Fuzzy equations allow delocalized collision. We will see that this delocalization offers a wider range of analytical tools and provides additional structure that can be exploit to show smoothness and global existence. I will also show examples of new Lyapunov functions of Fisher type.
Title: Fluctuations around the mean-field limit for Coulomb/Riesz gases
Abstract: This talk will report on recent developments in the description of log/Coulomb/Riesz gases beyond the leading order mean-field description, focusing on central limit theorems for fluctuations and asymptotics of correlation functions in varying temperature regimes. We focus on techniques based on the modulated (free) energy and estimates for their variations, known as commutator estimates, which have been successfully used to analyze the mean-field limit and may also be used to prove convergence of the fluctuations. The talk will highlight an important interplay between the initial size of fluctuations, the singularity of the interaction, and the scaling of the temperature that arises in analyzing the fluctuations. Time permitting, we will discuss both the equilibrium and non-equilibrium cases. Based on joint work with Jiaoyang Huang and Sylvia Serfaty.
Title: Carleman-Weighted Global Reconstruction for 3D Inverse Scattering with Partial Boundary Data.
Abstract: Inverse scattering problems arise in many applications where one wants to locate and characterize hidden objects from wave measurements, for example, in medical imaging, non-destructive testing, or subsurface exploration. This talk focuses on a three-dimensional frequency-domain scattering model governed by the Helmholtz equation with an unknown dielectric coefficient. The available data consist of multi-frequency backscattered measurements on one-sixth of the boundary generated by a single point source. We first describe a reduction in the frequency variable. Using a logarithmic change of variables and an expansion in a suitable polynomial–exponential (generalized Fourier) basis, we transform the original equation, which depends on both space and frequency, into a finite system of coupled quasi-linear elliptic equations that depend only on the spatial variables. To solve this nonlinear system, we apply the Carleman convexification method. Based on a weighted Carleman estimate, we construct a Tikhonov-type functional that is globally strictly convex on bounded sets. This structure proves that a simple gradient descent algorithm converges globally to the unique minimizer of the functional, regardless of the initial guess. The talk concludes with three-dimensional numerical reconstructions from both simulated and experimental data. These examples demonstrate that the method can accurately recover the locations and shapes of multiple types of scatterers.
Title: From kinetic equations of reactive Boltzmann type to the cross-diffusion system with Michaelis-Menten kinetics
Abstract: Reactions involving enzymes are fundamental in biochemistry, where enzymes act as catalysts in the reaction process. In the context of ODE models, one of the most widely used mechanisms for describing enzyme-catalyzed reactions is the Michaelis–Menten kinetics. This mechanism corresponds to a fast-reaction limit as the so-called QSS parameter ε ↘ 0. We present the rigorous derivation of the Michaelis–Menten kinetics in a macroscopic PDE setting, which was first proved in [1]. In particular, we show that a reaction–diffusion system describing an enzyme reaction governed by the law of mass action can be reduced, in the limit ε → 0, to a cross-diffusion system with Michaelis–Menten kinetics. Then, we introduce both the formal and rigorous derivation from a mesoscopic description, given by kinetic equations of reactive Boltzmann type, to the aforementioned cross-diffusion system (currently under study in [2]). This is joint work with Andrea Bondeasan.
[1] Quoc Bao Tang, BNT, Rigorous derivation of Michaelis-Menten kinetics in the presence of slow diffusion, SIAM Journal on Mathematical Analysis, Volume 56, Issue 5, 2024.
[2] Andrea Bondesan, BNT, From kinetic equations of reactive Boltzmann type to the crossdiffusion system with Michaelis-Menten kinetics, in Progress.
Title: Number Theory helps PDE Theory
Abstract: In two example cases we show how number theoretical insight can help develop new results in PDE theory. The first example deals with stability theory for port-Hamiltonian systems and the second with well-posedness for sign-indefinite divergence form problems.
Title: TBA
Abstract: TBA
Title: TBA
Abstract: TBA
Title: TBA
Abstract: TBA
Title: TBA
Abstract: TBA