Searcharxiv⌕ Search

arXiv subjects

Xiaoyu Zeng

Publications and source records attributed to Xiaoyu Zeng.

At least 19 recordsLinked to original sources
arXiv2026

Uniqueness, Nondegeneracy, and Asymptotic Expansions of Ground States for the Dirac-Coulomb System

We study ground states of the following Dirac-Coulomb system $$\mathscr{D}_cu_c- \left(\abs{x}^{-1}\ast \abs{u_c}^2\right)u_c=ω_cu_c,$$ under the constraint $\norm[L^2]{u_c}^2=1.$For sufficiently large \(c\), we prove uniqueness of ground states modulo translations, phase transformations and spinorial rotations, and determine the kernel of the linearized operator at ground state. We also obtain asymptotic expansions of ground state in the nonrelativistic limit regime.

math.AP↗
arXiv2026

Mountain-Pass Solutions for Second-Order Ergodic Mean-Field Game Systems

We study the existence of mountain-pass solutions to a potential-free mean-field game system in the whole space $\mathbb R^n$ under the mass-supercritical regime, assuming an aggregating local coupling and a $C^2$ Hamiltonian that is $γ$-homogeneous with $γ> 1$. Due to the lack of smoothness of the underlying variational structure, the standard deformation lemma and the classical mountain-pass theorem are not directly applicable. To overcome this difficulty, we constrain the nonlinear term and employ a two-stage linearization argument to establish the existence of least-energy solutions to an auxiliary mean-field game problem with general coercive potentials. In the vanishing coercive potential limit, we recover compactness by using maximal regularity for Hamilton-Jacobi equations together with Pohozaev-type identities, and show that the potential-free mean-field game system admits a classical solution, which is also an optimizer of a Gagliardo-Nirenberg type inequality. Finally, we analyze the mountain-pass geometry of the variational structure, which yields that the solution obtained above corresponds to a mountain-pass type solution of the original mean-field game system. These results provide an affirmative answer to the longstanding problem concerning the existence of mountain-pass solutions for mean-field game systems. Furthermore, as a byproduct, we relax the admissible set and provide a unified framework for establishing the optimal Gagliardo-Nirenberg inequality below the Sobolev critical exponent.

math.FA↗
arXiv2025

Existence and local uniqueness of multi-spike solutions for Brézis-Nirenberg problem with prescribed mass

In this paper, we consider the following Brézis-Nirenberg problem with prescribed $ L^2$-norm (mass) constraint: \begin{equation*} \begin{cases} -Δu=|u|^{2^*-2} u +λ_ρu\quad \text { in } Ω, u>0, \quad u \in H_0^1(Ω), \quad \int_Ω u^2dx=ρ, \end{cases} \end{equation*} where $N \geqslant 6$, $2^*=2 N /(N-2)$ is the critical Sobolev exponent, $ρ>0$ is a given small constant and $λ_ρ>0$ acts as an Euler-Lagrange multiplier. For any $k\in \mathbb{R}^+$, we construct a $k$-spike solutions in some suitable bounded domain $Ω$. Our results extend those in \cite{BHG3,DGY,SZ}, where the authors obtained one or two positive solutions corresponding to the (local) minimizer or mountain pass type critical point for the energy functional of above equation. Furthermore, using blow-up analysis and local Pohozaev identities arguments, we prove that the $k$-spike solutions are locally unique. Compared to the standard Brézis-Nirenberg problem without the mass constraint, an additional difficulty arises in estimating the error caused by the differences in the Euler-Lagrange multipliers corresponding to different solutions. We overcome this difficulty by introducing novel observations and estimates related to the kernel of the linearized operators.

math.AP↗
arXiv2025

Gagliardo-Nirenberg-Sobolev inequalities and ground states of Fermions in relativistic Hartree-Fock model

This paper presents a rigorous mathematical analysis of the relativistic Hartree-Fock model for finite Fermi systems. We first establish an optimal Gagliardo-Nirenberg-Sobolev (GNS) inequality with Hartree-type nonlinearities for orthonormal systems and characterize the qualitative properties of its optimizers. Furthermore, we derive a finite-rank Lieb-Thirring inequality involving convolution terms and show that it is the duality of the GNS-inequality-a result that, to our knowledge, has not previously appeared in the literature. For the relativistic Hartree-Fock model, we prove that ground states exist if and only if the coupling parameter $K<\mathcal{K}_\infty^{(N)}$, where $\mathcal{K}_\infty^{(N)}$ is the optimal constant in the GNS-inequality. Finally, under suitable assumptions on the external potentials, we calculate the precisely asymptotic behavior of ground states as $K\nearrow\mathcal{K}_\infty^{(N)}$.

math-ph↗
arXiv2025

Existence and Blow-up Profiles of Ground States in Second Order Multi-population Mean-field Games Systems

In this paper, we utilize the variational structure to study the existence and asymptotic profiles of ground states in multi-population ergodic Mean-field Games systems subject to some local couplings with mass critical exponents. Of concern the attractive and repulsive interactions, we impose some mild conditions on trapping potentials and firstly classify the existence of ground states in terms of intra-population and interaction coefficients. Next, as the intra-population and inter-population coefficients approach some critical values, we show the ground states blow up at one of global minima of potential functions and the corresponding profiles are captured by ground states to potential-free Mean-field Games systems for single population up to translations and rescalings. Moreover, under certain types of potential functions, we establish the refined blow-up profiles of corresponding ground states. In particular, we show that the ground states concentrate at the flattest global minima of potentials.

math.FA↗
arXiv2025

Local Minimizers in Second Order Mean-field Games Systems with Choquard Coupling

Mean-field Games systems (MFGs) serve as paradigms to describe the games among a huge number of players. In this paper, we consider the ergodic Mean-field Games systems in the bounded domain with Neumann boundary conditions and the decreasing Choquard coupling. Our results provide sufficient conditions for the existence of solutions to MFGs with Choquard-type coupling. More specifically, in the mass-subcritical and critical regimes, the solutions are characterized as global minimizers of the associated energy functional. In the case of mass supercritical exponents, up to the Sobolev critical threshold, the solutions correspond to local minimizers. The proof is based on variational methods, in which the regularization approximation, convex duality argument, elliptic regularity and Hardy-Littlewood-Sobolev inequality are comprehensively employed.

math.FA↗
arXiv2024

Blow-up Behaviors of Ground States in Ergodic Mean-field Games Systems with Hartree-type Coupling

In this paper, we investigate the concentration behaviors of ground states to stationary Mean-field Games systems (MFGs) with the nonlocal coupling in $\mathbb R^n$, $n\geq 2.$ With the mass critical exponent imposed on Riesz potentials, we first discuss the existence of ground states to potential-free MFGs, which corresponds to the establishment of Gagliardo-Nirenberg type's inequality. Next, with the aid of the optimal inequality, we classify the existence of ground states to stationary MFGs with Hartree-type coupling in terms of the $L^1$-norm of population density defined by $M$. In addition, under certain types of coercive potentials, the asymptotics of ground states to ergodic MFGs with the nonlocal coupling are captured. Moreover, if the local polynomial expansions are imposed on potentials, we study the refined asymptotic behaviors of ground states and show that they concentrate on the flattest minima of potentials.

math.FA↗
arXiv2024

Critical Mass Phenomena and Blow-up behavior of Ground States in stationary second order Mean-Field Games systems with decreasing cost

This paper is devoted to the study of Mean-field Games (MFG) systems in the mass critical exponent case. We firstly establish the optimal Gagliardo-Nirenberg type inequality associated with the potential-free MFG system. Then, under some mild assumptions on the potential function, we show that there exists a critical mass $M^*$ such that the MFG system admits a least energy solution if and only if the total mass of population density $M$ satisfies $M<M^*$. Moreover, the blow-up behavior of energy minimizers are captured as $M\nearrow M^*$. In particular, given the precise asymptotic expansions of the potential, we establish the refined blow-up behavior of ground states as $M\nearrow M^*.$ While studying the existence of least energy solutions, we establish new local $W^{2,p}$ estimates of solutions to Hamilton-Jacobi equations with superlinear gradient terms.

math.AP↗
arXiv2024

StyleSeg V2: Towards Robust One-shot Segmentation of Brain Tissue via Optimization-free Registration Error Perception

One-shot segmentation of brain tissue requires training registration-segmentation (reg-seg) dual-model iteratively, where reg-model aims to provide pseudo masks of unlabeled images for seg-model by warping a carefully-labeled atlas. However, the imperfect reg-model induces image-mask misalignment, poisoning the seg-model subsequently. Recent StyleSeg bypasses this bottleneck by replacing the unlabeled images with their warped copies of atlas, but needs to borrow the diverse image patterns via style transformation. Here, we present StyleSeg V2, inherited from StyleSeg but granted the ability of perceiving the registration errors. The motivation is that good registration behaves in a mirrored fashion for mirrored images. Therefore, almost at no cost, StyleSeg V2 can have reg-model itself "speak out" incorrectly-aligned regions by simply mirroring (symmetrically flipping the brain) its input, and the registration errors are symmetric inconsistencies between the outputs of original and mirrored inputs. Consequently, StyleSeg V2 allows the seg-model to make use of correctly-aligned regions of unlabeled images and also enhances the fidelity of style-transformed warped atlas image by weighting the local transformation strength according to registration errors. The experimental results on three public datasets demonstrate that our proposed StyleSeg V2 outperforms other state-of-the-arts by considerable margins, and exceeds StyleSeg by increasing the average Dice by at least 2.4%.

cs.CV↗
arXiv2024

Co-existence of Type II blow-ups with multiple blow-up rates for five-dimensional heat equation with critical nonlinear boundary conditions

We consider the following five-dimensional heat equation with critical boundary condition \begin{equation*} \partial_t u=Δu \mbox{ \ in \ } \mathbb{R}_+^5\times (0,T) , \quad -\partial_{x_5}u =|u|^\frac{2}{3}u \mbox{ \ on \ } \pp \mathbb{R}^5_+ \times (0,T) . \end{equation*} Given $\mathfrak{o}$ distinct boundary points $q^{[i]} \in \partial \mathbb{R}_+^5$, and $\mathfrak{o}$ integers $l_i\in \mathbb{N}$ (possibly duplicated), $i=1,2,\dots, \mathfrak{o}$, for $T>0$ sufficiently small, we construct a finite-time blow-up solution $u$ with a type II blow-up rate $(T-t)^{-3l_i -3}$ for $x$ near $q^{[i]}$. This seems to be the first result of the co-existence of type II blowups with different blow-up rates. To accommodate highly unstable blowups with different blowup rates, we first develop a unified linear theory for the inner problem with more time decay in the blow-up scheme through restriction on the spatial growth of the right-hand side, and then use vanishing adjustment functions for deriving multiple rates at distinct points. This paper is inspired by [25, 52, 60].

math.AP↗
arXiv2024

Infinitely many solutions for a class of fractional Schrodinger equations coupled with neutral scalar field

We study the fractional Schrödinger equations coupled with a neutral scalar field $$ (-Δ)^s u+V(x)u=K(x)ϕu +g(x)|u|^{q-2}u, \quad x\in \mathbb{R}^3,\qquad (I-Δ)^t ϕ=K(x)u^2, \quad x\in \mathbb{R}^3, $$ where $(-Δ)^s$ and $(I-Δ)^t$ denote the fractional Laplacian and Bessel operators with $\frac{3}{4} <s<1$ and $0<t<1$, respectively. Under some suitable assumptions for the external potentials $V$, $K$ and $g$, given $q\in(1,2)\cup(2,2_s^*)$ with $2_s^*:= \frac{6}{3-2s}$, with the help of an improved Fountain theorem dealing with a class of strongly indefinite variational problems approached by Gu-Zhou [Adv. Nonlinear Stud., {\bf 17} (2017), 727--738], we show that the system admits infinitely many nontrivial solutions.

math.AP↗
arXiv2023

On the Kirchhoff equation with prescribed mass and general nonlinearities

In the present paper, we apply a global branch approach to study the existence, non-existence and multiplicity of positive normalized solutions $(λ_c, u_c)\in \mathbb{R}\times H^1(\mathbb{R}^N)$ to the following Kirchhoff problem $$ -\left(a+b\int_{\mathbb{R}^N}|\nabla u|^2dx\right)Δu+λu=g(u)~\hbox{in}~\mathbb{R}^N,\;N\geq 1 $$ satisfying the normalization constraint $ \displaystyle\int_{\mathbb{R}^N}u^2=c, $ which appears in free vibrations of elastic strings. The parameters $a,b>0$ are prescribed as well as the mass $c>0$. Due to the presence of the non-local term $b\int_{\mathbb{R}^N}|\nabla u|^2dx Δu$, such problems lack the mountain pass geometry in the higher dimension case $N\geq 5$. Our result seems to be the first attempt in this aspect.

math.AP↗
arXiv2022

Robust One-shot Segmentation of Brain Tissues via Image-aligned Style Transformation

One-shot segmentation of brain tissues is typically a dual-model iterative learning: a registration model (reg-model) warps a carefully-labeled atlas onto unlabeled images to initialize their pseudo masks for training a segmentation model (seg-model); the seg-model revises the pseudo masks to enhance the reg-model for a better warping in the next iteration. However, there is a key weakness in such dual-model iteration that the spatial misalignment inevitably caused by the reg-model could misguide the seg-model, which makes it converge on an inferior segmentation performance eventually. In this paper, we propose a novel image-aligned style transformation to reinforce the dual-model iterative learning for robust one-shot segmentation of brain tissues. Specifically, we first utilize the reg-model to warp the atlas onto an unlabeled image, and then employ the Fourier-based amplitude exchange with perturbation to transplant the style of the unlabeled image into the aligned atlas. This allows the subsequent seg-model to learn on the aligned and style-transferred copies of the atlas instead of unlabeled images, which naturally guarantees the correct spatial correspondence of an image-mask training pair, without sacrificing the diversity of intensity patterns carried by the unlabeled images. Furthermore, we introduce a feature-aware content consistency in addition to the image-level similarity to constrain the reg-model for a promising initialization, which avoids the collapse of image-aligned style transformation in the first iteration. Experimental results on two public datasets demonstrate 1) a competitive segmentation performance of our method compared to the fully-supervised method, and 2) a superior performance over other state-of-the-art with an increase of average Dice by up to 4.67%. The source code is available at: https://github.com/JinxLv/One-shot-segmentation-via-IST.

cs.CV↗
arXiv2022

Uniqueness of Single Peak Solutions for Coupled Nonlinear Gross-Pitaevskii Equations with Potentials

For a couple of singularly perturbed Gross-Pitaevskii equations, we first prove that the single peak solutions, if they concentrate on the same point, are unique provided that the Taylor's expansion of potentials around the concentration point is in the same order along all directions. Among other assumptions, our results indicate that the peak solutions obtained in [21,31,38] are unique. Moreover, for the radially symmetric ring-shaped potential, which attains its minimum at the spheres$Γ_i:=\{x\in\mathbb{R}^N:|x|=A_i>0\},i=1,2,\cdots,l,$ and is totally degenerate in the tangential space of $Γ_i$, we prove that the positive ground state is cylindrically symmetric and is unique up to rotations around the origin. Aa far as we know, this is the first uniqueness result for ground states under radially symmetric but non-monotonic potentials.

math.AP↗
arXiv2022

A constrained minimization problem related to two coupled pseudo-relativistic Hartree equations

We are concerned with the following constrained minimization problem: $$e(a_{1},a_{2},β) := \inf\left\{E_{a_{1},a_{2},β}(u_{1},u_{2}): \|u_{1}\|_{L^{2}(\mathbb{R}^{3})} = \|u_{2}\|_{L^{2}(\mathbb{R}^{3})} = 1\right\},$$ where $E_{a_{1},a_{2},β}$ is the energy functional associated to two coupled pseudo-relativistic Hartree equations involving three parameters $a_{1}, a_{2}, β$ and two trapping potentials $V_1(x)$ and $V_2(x)$. In this paper, we obtain the existence of minimizers of $e(a_{1},a_{2},β)$ for possible $a_{1}, a_{2}$ and $β$ under suitable conditions on the potentials, which generalizes the results of the papers [16,17,18] in different senses.

math.AP↗
arXiv2021

Asymptotic behavior of least energy solutions for a fractional Laplacian eigenvalue problem on $R^N$

We are interested in the existence and asymptotical behavior for the least energy solutions of the following fractional eigenvalue problem \begin{equation*} (P)\quad (-Δ)^{s}u+V(x)u=μu+am(x)|u|^{\frac{4s}{N}}u,\quad \int_{\mathbb{R}^{N}}|u|^{2}dx=1,\ u\in H^{s}(\mathbb{R}^{N}), \end{equation*} where $s\in(0,1)$, $μ\in\mathbb{R}$, $a>0$, $V(x)$ and $m(x)$ are $L^{\infty}(\mathbb{R}^{N})$ functions with $N\geq2$. We prove that there is a threshold $a_s^{*}>0$ such that problem $(P)$ has a least energy solution $u_{a}(x)$ for each $a\in(0,a_s^{*})$ and $u_{a}$ blows up, as $a\nearrow a_s^{*}$, at some point $x_0 \in \mathbb{R}^N$, which makes $V(x_0)$ be the minimum and $m(x_0)$ be the maximum. Moreover, the precise blowup rates for $u_a$ are obtained under suitable conditions on $V(x)$ and $m(x)$.

math.AP↗
arXiv2020

Vision Skills Needed to Answer Visual Questions

The task of answering questions about images has garnered attention as a practical service for assisting populations with visual impairments as well as a visual Turing test for the artificial intelligence community. Our first aim is to identify the common vision skills needed for both scenarios. To do so, we analyze the need for four vision skills---object recognition, text recognition, color recognition, and counting---on over 27,000 visual questions from two datasets representing both scenarios. We next quantify the difficulty of these skills for both humans and computers on both datasets. Finally, we propose a novel task of predicting what vision skills are needed to answer a question about an image. Our results reveal (mis)matches between aims of real users of such services and the focus of the AI community. We conclude with a discussion about future directions for addressing the visual question answering task.

cs.HC↗