Peer Revenue Play credits. Real papers.

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Mathematics

math Watch Mathematics (sign in to use)

Sunday 26 April

Numerical Analysis

A positivity preserving and entropy stable nodal discontinuous Galerkin scheme for ideal MHD

Numerically solving magnetohydrodynamic (MHD) equations faces many challenges: avoiding divergence error, maintaining positivity, and satisfying entropy conditions. Among discontinuous Galerkin (DG) schemes, there has been a modal version that is locally divergence-free and positivity preserving and a nodal version that is semi-discretely entropy stable. In this work, we develop a DG scheme that combines the advantages of these two and solves all the three challenges. The key ingredients that bring these two schemes together are an HLL numerical flux with entropy stable signal speed estimates and a locally divergence-free projection. To handle problems with strong shocks, the essentially oscillation-free damping is applied. Various numerical experiments verify the accuracy and robustness of our method.
Read the full abstract

Probability

Sharp pathwise nonuniqueness for additive SDEs

We construct a family of velocity fields demonstrating the sharpness of the classical Zvonkin–Veretennikov–Davie strong well-posedness by noise regime. We consider stochastic differential equations driven by Brownian noise with drift $u$ and show that for any $\alpha<0$, there exists a velocity field $u \in L^\infty_t C^\alpha_x$ that admits a unique weak solution but does not satisfy pathwise uniqueness (and hence has no strong solutions). This contrasts with the case $\alpha \geq 0$, for which the existence of a unique strong solution is guaranteed. The velocity field construction is random, and the proof essentially uses central limit theorem scaling through the Berry–Esseen theorem. We also give natural extensions to non-Brownian driving noises, including nonuniqueness for arbitrary driving noises with certain Hölder regularities and an analogous sharpness of the strong well-posedness by noise regime for fractional Brownian motions.
Read the full abstract

Combinatorics

Pair-Trace Absorption Certificates for Regular Induced Subgraphs

We study a fixed-core absorption problem for regular induced subgraphs. A set is q-modular if all induced degrees are congruent modulo q. Given a q-modular witness A and a retained core U subset A, we ask when deleting equal-trace q-tuples from A\U can make U into a 2q-modular witness. The main contribution is a finite absorption-or-obstruction certificate. We give an exact quotient formula for the deletion-tail obstruction in complement-orbit coordinates: the correct expression uses oriented differences n_B - n_{U\B}, not sums. Equal-trace q-tuples absorb exactly the span of their trace classes in F_2^U / 1_U. In particular, a connected graph of q-heavy two-point traces on U, together with one odd trace when |U| is even, absorbs every top-bit defect by deleting at most q(|U|-1) tail vertices. If fixed-core absorption fails, the obstruction is an explicit even parity cut of U. We also record the parity base, the terminal modular criterion, and a conditional modular-witness threshold theorem explaining the relevance to the Erdos-Fajtlowicz-Staton problem. The paper does not claim to solve that problem or to improve the general lower bound for F(n).
Read the full abstract

Dynamical Systems · Cross-listed from physics.flu-dyn

Deep Learning of Solver-Aware Turbulence Closures from Nudged LES Dynamics

The differentiable physics paradigm may be leveraged as an a-posteriori approach for discovering turbulence closure models by embedding a neural network parameterization directly inside the solver and optimizing it given potentially sparse target data. This addresses a key limitation of a-priori learning where direct numerical simulation (DNS) data is used to approximate the subgrid stress with the assumption of a low-pass filter. Closures trained in this a-priori manner frequently lead to unstable deployments due to the mismatch between the assumed filter and the effect of numerical discretizations and coarse-graining. In comparison, while typically stable during deployment, a-posteriori learning incurs high computational costs due to the need to backpropagate through a large eddy simulation (LES) solver. Furthermore, a-posteriori methods are challenging to apply broadly since they require significant modification of existing solvers. Finally, both approaches are limited when generalization is desired across different numerical schemes with their implicit filtering characteristics. In this work, we present a deep-learning approach for turbulence closure modeling built on the continuous data assimilation framework. Our approach enables the a-priori training of closures using sparsely observed DNS data without modifying or differentiating through the LES solver, while preserving stability during deployment for the recovery of invariant statistics. We focus on the model's ability to adapt to different discretizations by explicitly conditioning it on the numerical scheme. We use two- and three-dimensional canonical cases to test our framework and show that the learned correction systematically tracks the discretization error of the coarse solver.
Read the full abstract

Algebraic Geometry · Cross-listed from cs.SC

Enhanced CAD-Based Quantifier Elimination With Multiple Equational Constraints

This paper presents two enhancements to cylindrical algebraic decomposition (CAD) based quantifier elimination (QE) for cases in which multiple equational constraints are present in the given input formula $\phi^*$. The first enhancement provides more detail in the output when there is a conceptual partition of the set of variables of $\phi^*$ into parameters and unknowns. In such cases, we describe how to partition the parameter space so that: (1) in each open set of the partition the number $\nu$ of associated unknowns is a finite constant or is infinite; and (2) for each such open set for which $\nu$ is finite, an expression for the unknowns in terms of the parameters is provided. The second enhancement is an efficiency gain achievable in certain situations. Indeed, when certain conditions are met, the second CAD equational projection step can be reduced more significantly than is supported by the prior existing theory. Relevant theorems and worked examples for both enhancements are provided. Application areas include approximation theory, cuspidal manipulator classification, and biological/chemical systems.
Read the full abstract

Algebraic Geometry

The convolution algebra of constructible sheaves

Let \(E\) be a finite-dimensional real vector space. We study invertible objects in the monoidal category of constructible sheaves on \(E\), endowed with the convolution product \(\star\). We show that the inverse of an invertible constructible sheaf \(F\) is the dual of its antipodal transform. We also prove that a compactly supported constant sheaf is invertible if and only if its support is convex. We also introduce a microlocal transform \(B(F)\), obtained by projecting the characteristic cycle of $F$ to \(E^*\), and prove that it is compatible with convolution. This yields a necessary condition for invertibility.
Read the full abstract

Number Theory

On optimization on ravine functions. Minkowski-Cohn moduli surface in Cohn parameterization

This paper presents a brief overview of ravine functions using the example of the Minkowski-Cohn moduli surface from the point of view of optimization on it. Elements of representation and solution of the minimization problem at a point are presented.

Functional Analysis

Bourgain's method for K-closedness in the semicommmutative setting

In the early 1990s, J.Bourgain proved a general result $K$-closedness result in the context of classical harmonic analysis. In this paper, we extend Bourgain's method to the semicommutative setting, making use of the recent semicommutative Calderón-Zygmund decomposition introduced by L.Cadilhac, JM.Conde-Alonso and J.Parcet. As an application, we recover Pisier's result about $K$-closedness of noncommutative Hardy spaces on the torus, and we also establish new interpolation results for noncommutative Sobolev spaces on the torus.
Read the full abstract

Mathematical Physics · Cross-listed from physics.flu-dyn

Dissipative Vortex Binaries in Compact Fluid Domains with Geometric Corrections

We study a dissipative extension of vortex-binary motion in a doubly periodic fluid domain. The underlying conservative system admits an exact integrable reduction to a single complex relative coordinate. Dissipation is introduced via a minimal rotated-velocity (mutual-friction) term, as motivated by finite-temperature superfluid dynamics, converting the Hamiltonian evolution into a mixed symplectic–gradient flow with monotonic energy decay for quantized vortices. In the local regime, the dissipative binary remains analytically solvable and admits closed-form solutions, with systematic corrections arising from the toroidal geometry. Equal same-sign vortices execute outward spiraling motion, while equal opposite-sign pairs (dipoles) undergo finite-time collapse in the planar limit. On the torus, however, the dipole orientation is no longer invariant: the geometry induces a slow angular drift, even in regimes where planar dynamics would preserve alignment. For unequal opposite-sign pairs, dissipation induces coupled contraction and rotation, leading to a finite-time nonlinear chirp characterized by $\dot{\omega}\propto\omega^2$, in contrast with electromagnetic and gravitational inspirals where $\dot{\omega}\propto \omega^{3}$ and $\dot{\omega}\propto \omega^{11/3}$. These results highlight the interplay between Hamiltonian structure, dissipation, and geometry in periodic fluid systems.
Read the full abstract

Analysis of PDEs

Very weak solutions of the heat equation with anisotropically singular time-dependent diffusivity

We investigate the heat equation with a time-dependent, anisotropic, and potentially singular diffusivity tensor. Since weak (in the Sobolev sense) or distributional solutions may not exist in this setting, we employ the framework of very weak solutions to establish the existence and uniqueness of solutions to the heat equation with singular, anisotropic, time-dependent diffusivity.
Read the full abstract

All authors Zhirayr Avetisyan, Zahra Keyshams, Monire Mikaeili Nia, Michael Ruzhansky

Spectral Theory

Generalized Aubry-André formula and continuity of the intersection spectrum of the Almost Mathieu operator

We consider the spectrum of the Almost Mathieu operator (AMO) and show that the moments of the restriction of the Lebesgue measure to the intersection spectrum $\text{Leb}|_{\Sigma_{\alpha,\lambda}}$ are polynomials in coupling $\lambda$ with coefficients that are trigonometric polynomials in frequency $\alpha$. The statement can be considered as a generalization of the Aubry-André formula for the measure of the spectrum of AMO. As a corollary, we obtain that the restriction of the Lebesgue measure to the intersection spectrum that we denote by $\mu^{-}_{\alpha, \lambda}$ depends continuously on the parameters (frequency $\alpha$ and coupling $\lambda$) in weak-* topology. Moreover, we prove that the dependence is not just continuous but analytic in $\lambda$ and $C^{\infty}$ in $\alpha$ in a sense that an integral of an analytic test function $\varphi(x)$ with respect to $\mu^{-}_{\alpha, \lambda}$ has the same kind of dependence. In particular, this implies that the Lebesgue measure of the part of the spectrum $\Sigma_{\alpha,\lambda}$ that lies between two gaps depends analytically on the coupling constant $\lambda$ and $C^{\infty}$ on the frequency $\alpha$ in an open domain (away from the critical coupling $\lambda=1$) where these gaps do not bifurcate.
Read the full abstract

Statistics Theory · Cross-listed from stat.ME

Bayesian change-plane regression

Change-plane regression represents treatment effect heterogeneity through an interpretable rule that assigns patients to two groups according to whether a linear score of baseline covariates crosses a threshold. Although easy to communicate, likelihood-based inference for the hard threshold is nonregular, invalidating standard large-sample approximations. We develop a Bayesian framework that replaces the sharp indicator with a smooth probit gate at a declared smoothing scale and treats the smoothed subgroup rule as the reported estimand. At a fixed smoothing scale, the posterior satisfies a misspecified Bernstein–von Mises theorem centered at the smoothed rule. Under a vanishing smoothing schedule, the posterior learns the boundary faster than the parametric rate, while within an explicit schedule window the regression block and treatment-effect contrast satisfy a Bernstein–von Mises theorem centered at the hard-threshold values with known-subgroup oracle efficiency. Thus, effect inference pays no first-order price for regularization. A decision-theoretic reporting protocol with consistency guarantees separates evidence of heterogeneity from reporting a boundary. Computation uses latent-variable augmentation, a great-circle elliptical slice sampler, and a normalized horseshoe prior. Simulations and an application to the PREMIER trial illustrate the method.
Read the full abstract

Symplectic Geometry

Contact flexibility and rigidity for toric Gorenstein prequantizations and Ehrhart theory of toric diagrams

Gorenstein toric contact manifolds are good toric contact manifolds with zero first Chern class that are completely determined by certain integral convex polytopes called toric diagrams. The Ehrhart polynomial of these toric diagrams determines and is determined by the contact Betti numbers of the corresponding contact manifolds, i.e. the dimension of their cylindrical contact homology in eachdegree. In this paper we look into the following natural question: to what extent do these contact invariants determine the Gorenstein toric contact manifold? Flexibility is the norm and we illustrate it with the family of Gorenstein toric contact manifolds that arise as the prequantization of monotone iterated ${\mathbb P}^1$-bundles, i.e. monotone Bott manifolds. In each dimension, the Ehrhart polynomial of their toric diagrams is equal to the Ehrhart polynomial of the cross-polytope, corresponding to the monotone prequantization of ${\mathbb P}^1 \times \cdots \times {\mathbb P}^1$, and we describe the unimodular classification of these toric diagrams. On the rigidity side, we will show that the primitive prequantization of ${\mathbb P}^1 \times \cdots \times {\mathbb P}^1$ is rigid, i.e. completely determined as a Gorenstein toric contact manifold by its contact Betti numbers. More precisely, in each dimension, we show that its toric diagram, which we name small cross-polytope, is the unique toric diagram with its particular Ehrhart polynomial. We will also prove a rigidity result for a family of Gorenstein toric contact manifolds that arise as the primitive prequantization of monotone ${\mathbb P}^1$-bundles over ${\mathbb P}^{n-1}$.
Read the full abstract

Analysis of PDEs

Fine structure of the two-phase Bernoulli free boundaries in 2D

We prove that the branching set of a solution to a two-dimensional two-phase Bernoulli problem with constant coefficients is locally finite. We do this via a Weierstrass representation formula, which allows to transform the problem into a new geometric two-phase problem for capillary minimal surfaces. We also apply this method to the obstacle problem establishing a connection between the directional derivatives of solutions to the obstacle problem and the linear thin two-membrane problem.
Read the full abstract

Probability

Kac's walk on rotation matrices mixes in $n^2 \log n$ steps

Kac's walk on the rotation group, introduced by Hastings in 1970, is an important high-dimensional Markov chain with applications in statistical physics, statistics, cryptography, and computational science. Despite its simple transition rules, determining its total-variation mixing time has remained a challenging problem for decades. A key obstacle is that the walk is not conjugation-invariant, placing it beyond the reach of classical Fourier-analytic techniques that apply to many related random walks on compact groups. We prove that Kac's walk mixes in total variation in \(O(n^2 \log n)\) steps, matching the conjectured mixing time up to constants. The proof is based on a refined two-stage coupling. Building on earlier work, the first stage contracts two copies of the chain to a small neighborhood via a Wasserstein coupling. Our main contribution is a new framework for analyzing the second-stage coupling. It can be viewed as a discrete analogue of Malliavin calculus for Markov chains. We represent the law of the chain as the pushforward of high-dimensional noise and prove quantitative non-degeneracy of the associated linearization using matrix martingale methods. This yields an approximately Gaussian distribution in the Lie algebra with well-conditioned covariance, allowing small group translations to be absorbed at negligible cost in total variation. Our approach provides a general framework for studying mixing in high-dimensional Markov chains in continuous state spaces with singular transition kernels.
Read the full abstract

Combinatorics

Recursive Record Filtering and Longest Decreasing Subsequences

We consider a recursive record-filtering procedure, which we informally call Disappear-Sort, acting as a sort of parallel to traditional patience sorting. Let $D_n$ denote the number of passes required to eliminate a sequence of length $n$ sampled as i.i.d.\ copies of a continuous random variable, where each pass retains the left-to-right records and applies the same rule recursively to the remaining entries. For the non-resampling procedure, we associate to a permutation $p_n\in S_n$ a natural poset and show that the recursive Disappear-Sort layers form an antichain decomposition of this poset. This provides an order-theoretic interpretation of the procedure and identifies the total number of passes with $L(p_n)$, the length of the longest decreasing subsequence of $p_n$. Equivalently, after reversing the comparison direction, the Disappear-Sort layers coincide with the piles arising in classical patience sorting. For a uniformly random permutation, the pass count therefore has the same distribution as the first-column length of the tableau shape produced by the Robinson–Schensted correspondence. We use this classical connection to express $\mathbb{E}[D_n]$ as a sum over partitions and standard Young tableaux. Established results on Plancherel-random Young diagrams then imply $\mathbb{E}[D_n]\sim 2\sqrt{n}$, with fluctuations on the $n^{1/6}$ scale governed by the Tracy–Widom distribution. We also consider a resampling variant in which the nonrecord entries are replaced after each pass by a fresh independent sample of the same size, and derive an exact recurrence for its expected number of passes involving the unsigned Stirling numbers of the first kind. We conclude with an $O(n\log n)$ implementation for computing the non-resampling pass count.
Read the full abstract

Information Theory · Cross-listed from cs.NI

Optimizing Information Freshness for Wireless Local Area Networks with Multiple APs

Dense indoor WLANs increasingly rely on multiple access points (APs) operating over partially overlapping spectrum to support latency-sensitive applications. In such deployments, simultaneous transmissions across APs create co-channel and adjacent-channel interference, making scheduling decisions interdependent and directly impacting information freshness. Motivated by emerging software-defined WLAN architectures that enable centralized coordination, we study the problem of minimizing network-wide Age of Information (AoI) in multi-AP WLANs. Unlike classical AoI scheduling that runs at a single AP, each scheduling decision is now coupled across APs due to interference. This leads to a new class of combinatorial AoI control problems with action-dependent time evolution. We first derive a lower bound on the achievable AoI under arbitrary scheduling policies. We then design stationary randomized policies that have constant-factor optimality guarantees relative to this bound. Building on these insights, we develop a Lyapunov drift-based online policy for systems with action-dependent frame lengths, and establish constant-factor guarantees using new ratio-based drift analysis. To enable scalable implementation, we further show that per-frame scheduling admits efficient polynomial-time local-search approximations under a submodularity assumption. Simulations using realistic WLAN layouts demonstrate about 50% AoI reduction over distributed single AP baselines.
Read the full abstract

Probability

Finite-sample Borel–Cantelli inequalities under mixing conditions

We prove explicit one-lag, finite-$N$ lower bounds for $\mathbb P(\bigcup_{k=1}^{N}A_{k})$ that use only the marginal probabilities $\mathbb P(A_{k})$ and a single selected-lag dependence coefficient of the event-generated $\sigma$-fields. A residue-class blocking argument gives, under $\varphi$-mixing, a bound with a free spacing parameter $L\ge 0$, spacing constant $1/(L+1)$, and residual governed by $\varphi(L+1)$; a strong-mixing covariance argument gives an $\alpha$-mixing analogue with an additive residual $\lceil N/(L+1)\rceil\,\alpha(L+1)$. When the lag-$(L+1)$ coefficient vanishes, both reduce to the finite-sample $m$-dependent bound of Panraksa (2026), and the spacing constant $1/(L+1)$ is sharp in this zero-residual sense. A second-order Bonferroni refinement and a worked geometrically $\varphi$-mixing example are included. The estimates are non-asymptotic one-lag tools, complementary to second-moment and variance criteria rather than competitors to them.
Read the full abstract

Combinatorics

A note on The asymptotic uniform distribution of subset sums

We find out that the main result of the article The asymptotic uniform distribution of subset sums can be proven much more easily, using an explicit formula proposed by Li and Wan.

Number Theory

Unbounded logarithmic limsup in Erdős Problem 684 via shifted carry scheduling

For $1\leq k\leq n$, let $u(n,k)=\prod_{p\leq k}p^{\nu_p\binom nk}$ and $f(n)=\min\{1\leq k\leq n:u(n,k)>n^2\}$. The minimum is interpreted as $+\infty$ if the set is empty. Here $\nu_p(m)$ denotes the exponent of the prime $p$ in $m$. Erdős Problem 684 asks for bounds on $f(n)$. We prove $\limsup_{n\to\infty} \frac{f(n)}{\log n}\frac{\log\log\log n}{\log\log n}\geq\frac12$. In particular, $\limsup_{n\to\infty}f(n)/\log n=\infty$, so no uniform estimate $f(n)=O(\log n)$ is possible. The proof constructs integers $n=tL_M-h-1$, where $L_M=\operatorname{lcm}(1,\ldots,M)$. Product-cell coding yields simultaneous two-sided approximations to $tL_M$ modulo every prime in $(M,K]$. Writing $n+1=tL_M-h$, the shift by $h$ folds both signs into the same one-sided carry region. The carries at levels at most $M$ are bounded by $\log\binom{h+k}{h}$. For the prime powers above $K$, a truncated CRT witness has modulus below the search range; exponential weighting then gives an exponentially small exceptional set. An elementary anchored-fibre lemma selects a multiplier satisfying both requirements. Together, these ingredients prove the stated bound unconditionally. The theorem has been formally verified in Lean 4 with Mathlib.
Read the full abstract

2604.23784 · 26 Apr · 13 pages. v3: complete rewrite. The v1-v2 argument, which relied on an unjustified weighted extension of Timofeev's method, is withdrawn and replaced by an independent unconditional proof of the stronger quantitative bound. The theorem is formally verified in Lean 4 (Mathlib + PrimeNumberTheoremAnd), with no hypotheses: https://github.com/jidodat/erdos684-lean (tag v3-arxiv)

Operator Algebras

Banach-compact operators, $\mathcal A$-precompactness, and frames in Hilbert $C^*$-modules

For a pair of Hilbert $C^*$-modules $\mathcal{M}, \mathcal{N}$ over a $C^*$-algebra $\mathcal{A}$, there are two notions of “rank-one operators”: $\theta_{x,y} : \mathcal{M} \to \mathcal{N}$, $\theta_{x,y}(z) = x \langle y, z \rangle$, where $y,z \in \mathcal{M}$, $x \in \mathcal{N}$ (these operators are called elementary $\mathcal{A}$-compact or elementary Kasparov operators) and $\theta_{x,f} : \mathcal{M} \to \mathcal{N}$, $\theta_{x,f}(z) = x f(z)$, where \(z \in \mathcal{M}\), \(x \in \mathcal{N}\), and \(f\) is a bounded \(\mathcal{A}\)-functional on \(\mathcal{M}\) (introduced by Manuilov). They generate a \(C^*\)-bimodule \(\mathbf{K}(\mathcal{M}, \mathcal{N})\) (\(\mathcal{A}\)-compact operators) over the \(C^*\)-algebras of adjointable operators and a Banach bimodule \(\mathbf{BK}(\mathcal{M}, \mathcal{N})\) (Banach-compact operators) over the algebras of all bounded morphisms, respectively. In order to give a geometric characterization of these classes of operators, we consider the notion of \(\mathcal{A}\)-precompactness (developing the one introduced by Manuilov). Banach-compact operators can be characterized as those with \(\mathcal{A}\)-precompact image of the unit ball. Another characterization obtained here is in terms of total boundedness of this set with respect to the uniform structure introduced by one of us previously. The constructions and proofs turn out to be closely related to the concept of frames in a Hilbert $C^*$-module.
Read the full abstract

Numerical Analysis

Asymptotic preserving scheme for the shallow water equations with non-flat bottom topography and Manning friction term

In our previous work [29], we proposed a class of high-order asymptotic preserving (AP) finite difference weighted essentially non-oscillatory (WENO) schemes for solving the shallow water equations (SWEs) with bottom topography and Manning friction, utilizing a penalization technique inspired by [6]. Although the added weighted diffusive term enhanced stability, it increased computational cost and slowed down the convergence rate in the intermediate regime between convection and diffusion. In this paper, we extend our previous study by removing the penalization while preserving the AP property. To achieve this, we employ a high order semi-implicit implicit-explicit Runge-Kutta (SI-IMEX-RK) time discretization, coupled with the high-order WENO reconstruction for first-order derivatives and a central difference scheme for second-order spatial derivatives. This combination yields a class of fully high-order schemes. Theoretical analysis and numerical experiments demonstrate that the proposed schemes retain AP, asymptotically accurate (AA) and well-balanced properties, while offering higher computational efficiency compared to our previous schemes in [29], especially in the intermediate regime between convection and diffusion. Moreover, treating the momentum in the friction terms implicitly is essential for preserving the AP property; otherwise, the scheme fails to converge to the limiting equations. This indicates that implicit treatment of Manning friction is necessary for the stability of the method.
Read the full abstract

Functional Analysis · Cross-listed from cs.LG

Necessary and sufficient conditions for universality of Kolmogorov-Arnold networks

We analyze the universal approximation property of Kolmogorov-Arnold Networks (KANs) in terms of their edge functions. If these functions are all affine, then universality clearly fails. How many non-affine functions are needed, in addition to affine ones, to ensure universality? We show that a single one suffices. More precisely, we prove that deep KANs in which all edge functions are either affine or equal to a fixed continuous function $\sigma$ are dense in $C(K)$ for every compact set $K\subset\mathbb{R}^n$ if and only if $\sigma$ is non-affine. In contrast, for KANs with exactly two hidden layers, universality holds if and only if $\sigma$ is nonpolynomial. We further show that the full class of affine functions is not required; it can be replaced by a finite set without affecting universality. In particular, in the nonpolynomial case, a fixed family of five affine functions suffices when the depth is arbitrary. More generally, for every continuous non-affine function $\sigma$, there exists a finite affine family $A_\sigma$ such that deep KANs with edge functions in $A_\sigma\cup\{\sigma\}$ remain universal. We also prove that KANs with the spline-based edge parameterization introduced by Liu et al.~\cite{Liu2024} are universal approximators in the classical sense, even when the spline degree and knot sequence are fixed in advance.
Read the full abstract

Analysis of PDEs

A Fujita-type threshold for the semilinear damped wave equation with Hartree-type nonlinearity and initial data from homogeneous Besov spaces

In this paper, we consider the semilinear damped wave equation with Hartree-type nonlinearity $\mathcal{I}_\gamma\left(|u|^{p_1}\right)|u|^{p_2}$, where $0<\gamma<n$, with initial data possessing additional negative regularity in $\dot{B}_{2, \infty}^{-\beta}$, $\beta \in [0,n/2)$. We first establish decay estimates for the corresponding linear equation in homogeneous Besov spaces $\dot{B}_{q,r}^{s}$. A key point is that the low-frequency part of the data is measured only in $\dot{B}_{m,\infty}^{-\beta}$, while the heat-like behavior yields the required $\ell^r$ summability under the condition $s+\beta+n(1/m-1/q)>0$. This formulation is particularly suited to the nonlinear analysis. We then establish the existence of a unique global mild solution for sufficiently small initial data whenever $$ p_1+p_2\geq 1+\frac{4+2\gamma}{n+2\beta}, $$ under the remaining admissibility conditions stated in the existence theorem. In particular, the critical case is covered whenever the critical line satisfies these conditions. Conversely, for initial data satisfying an explicit positive lower bound, the test-function method rules out global weak solutions when $$ 2<p_1+p_2<1+\frac{4+2\gamma}{n+2\beta}. $$ Thus, whenever the critical line is admissible and the subcritical interval is nonempty, $1+(4+2\gamma)/(n+2\beta)$ gives a Fujita-type threshold for the sum $p_1+p_2$.
Read the full abstract

Optimization and Control

Regret-Optimal Control for Finite-State Systems

We study the control of finite-state systems driven by exogenous disturbances, and design causal policies that track the performance of a lookahead benchmark controller. This objective is formalized through dynamic regret, so that favorable disturbance sequences are compared against a strong benchmark, while under adverse disturbance sequences the comparison accounts for the benchmark's degraded performance. This benchmark-relative framework provides an alternative to classical MDP formulations, which assume i.i.d. disturbances, and to robust control approaches, which optimize against worst-case disturbances. Our main result is a nested dynamic-programming solution that computes both the optimal worst-case regret and a regret-optimal policy. In particular, we introduce the Regret-Bellman operator, whose fixed-point value function feeds into a finite-horizon dynamic program. Numerical examples show that regret-optimal policies interpolate nicely between MDP-based and robust controllers without requiring knowledge of the disturbance distribution, and can even outperform both under i.i.d. or structured disturbances.
Read the full abstract

Optimization and Control

A Retraction-Free EXTRA Method for Decentralized Optimization on the Stiefel Manifold

Decentralized optimization provides a fundamental framework for large-scale learning and signal processing with distributed data. We study decentralized optimization with orthogonality constraints on the Stiefel manifold and propose RF-EXTRA, a distributed retraction-free primal-dual method on static undirected networks. The method combines an approximate gradient mapping for orthogonality-constrained optimization with an EXTRA-based decentralized recursion, thereby avoiding retractions while preserving a simple communication pattern. On the theoretical side, the analysis considers \revise{the joint error} $(\mathbf{X}_k-\overline{\mathbf X}_k,\mathbf{s}_k-\overline{\mathbf s}_k)$ in the local variables and local directions, and establishes a contractive recursion for the joint error. This contractivity ensures that the joint error can be controlled using small yet constant step sizes, thus leading to an exact $\mathcal{O}(1/K)$ convergence rate of RF-EXTRA to a stationary point. Experiments on PCA and low-rank matrix completion show that RF-EXTRA compares favorably with the reported decentralized baselines and exhibits strong communication efficiency on the tested tasks on the Stiefel manifold.
Read the full abstract

Probability

Large deviation principles for pattern-avoiding permutations, and limit shapes for constrained Mallows permutations

We study Mallows random permutations conditioned to avoid a given pattern $\alpha$ of length~$3$. When the bias parameter is of the form $e^{\beta/n}$, we prove that these permutations converge to a non-trivial explicit deterministic permuton that depends on the pattern $\alpha$ and on the parameter $\beta$. Along the way, we provide parametrizations for $\alpha$-avoiding permutons, and establish a large deviation principle for uniform $\alpha$-avoiding permutations. As a byproduct of the proof, we also obtain asymptotic estimates of two versions of $q$-Catalan numbers in the regime $q=e^{\beta/n}$.
Read the full abstract

All authors Thomas Budzinski, Victor Dubach, Valentin Féray, Mohamed Slim Kammoun, Mylène Maïda

Optimization and Control

Enforcing TSP-Optimality in Fair Vehicle Routing by Cutting Planes

We study the fair capacitated vehicle routing problem, in which a fleet of vehicles must serve a set of customers such that the difference between the longest and shortest route, the range, is minimized. A key challenge is that the range objective is non-monotonic: it can be reduced by artificially lengthening routes, leading to solutions that violate TSP-optimality of individual routes. Existing exact methods struggle to handle this efficiently. We propose a branch-price-and-cut framework that enforces TSP-optimality through TSP-optimality cuts, which forbid TSP-dominated arc sequences. We strengthen the cuts through a dedicated lifting procedure. Computational experiments on benchmark instances with up to 25 customers show the method solves nearly all instances to optimality, achieving an average gap of 0.27% on the hardest configurations.
Read the full abstract

Combinatorics

Modular Schur numbers

We study modular analogues of Schur numbers for systems of linear equations. We show that these only depend on the number of equations, not their coefficients and in the case of one equation show stronger bounds.