Papers

We define cone-additive functions for random walks on free products of countable graphs. These functions satisfy a limit theorem under mild assumptions. In fact, cone-additivity is present in several well-studied notions, like entropy, asymptotic range and drift. Cone-additivity can be seen as a separation property by space – a quite different perspective than the well-studied concept of sub-additivity in the context of free products of groups, which is a separation by time. In our inhomogeneous setting of free products of graphs, this separation by space allows us to deduce new limit theorems for travelling salesman problems (that is, distance functions of lamplighter random walks on free products), for weight functions on edges and the range of the $r$-th visit.

For $k\geq 1$ and $n\geq 2$, we introduce the multi-welded twin group $M_kWT_n$, a natural welded analogue of the multi-virtual twin group. We show that $M_kWT_n$ arises naturally as a quotient of the universal welded braid group $UW_n(k)$, placing it within the unified framework of universal virtual and welded braid-type groups. We establish natural quotient maps relating $M_kWT_n$ to the multi-virtual twin group $M_kVT_n$, the welded twin group $WT_n$, and the corresponding virtual and welded braid-type groups. Several structural properties of $M_kWT_n$ are obtained. In particular, we compute its abelianization, prove that its commutator subgroup is perfect for $n\ge5$, and show that the symmetric group $S_n$ is its smallest non-abelian finite quotient. We also investigate the representation theory of $M_kWT_n$. In fact, we classify all non-trivial complex homogeneous $2$-local representations of $M_kWT_n$, showing that only one family survives under the additional twin and welded relations. Furthermore, we classify all non-trivial complex homogeneous $3$-local representations of $M_2WT_n$. We further investigate the reducibility and faithfulness properties of both the $2$-local and $3$-local representations.

Parameter-dependent dynamical systems that exhibit bifurcations pose significant computational challenges, as traditional continuation methods require repeated, costly simulations across large ranges of parameter values to capture sudden qualitative changes in the solution. In this work, we propose a systematic approach to reconstruct the branches of the entire bifurcation diagram in a single numerical solver leveraging generalized Polynomial Chaos (PC) expansion. By treating the parameter as a random variable, we cast the deterministic parameter-dependent model in a weak stochastic form, and then use a Galerkin projection to recover bifurcation branches globally across the parameter domain without iterative pointwise continuation. We show that the resulting Galerkin system, in the non-uniqueness regime, produces many discrete algebraic roots that naturally split into two classes: highly oscillatory solutions and branch-approximating ones. We develop a rigorous theoretical framework that establishes consistency, proves convergence of the branch-approximating solutions to the true steady states, and guarantees uniqueness of the Galerkin solution under suitable assumptions. Finally, we confirm these theoretical results with numerical experiments on several parameter-dependent ordinary differential equations (ODEs), demonstrating the accuracy and computational efficiency of our single-run framework in capturing complex bifurcation diagrams for both scalar and vector-valued systems.

In this paper, we propose an integro-differential model for the spatio-temporal evolution of infectious diseases with asymptomatic transmission. The model consists of a reaction-diffusion system with an integral memory term accounting for the distribution of the incubation period. We first analyze the asymptotic behavior and the properties of the integro-differential model. Then, we prove the local existence of a weak solution of the system by means of the Faedo-Galerkin method and a compactness argument. The model is applied to simulate the geographical evolution of a disease in Lebanon.

We provide a complete and optimal characterization of $\mathbb{Q}$-Gorenstein normal affine singularities in terms of their K-stability in the sense of Collins–Székelyhidi.

We prove two existence results about block gluing in two-dimensional SFTs. First, a large class of functions between exponential and logarithmic can be realized as block gluing functions of two-dimensional SFTs. Second, there exists an aperiodic linearly block gluing two-dimensional SFT with entropy dimension 1. These solve two open questions of Gangloff and Sablik.

We present a neural-network-based framework for the solution of three-dimensional boundary value problems where the solution is expressible in terms of harmonic potentials. The approach leverages the Whittaker integral formula, which allows representing the solution through functions that are holomorphic with respect to a suitable complex variable. These functions are subsequently approximated using holomorphic neural networks, which guaranty fulfillment of the holomorphicity requirement. A key feature of the proposed formulation is that the governing partial differential equations (PDEs) are satisfied exactly by construction. Therefore, in contrast to standard physics-informed neural networks, no residual minimization of PDEs is required in the interior of the domain, and training is based exclusively on boundary collocation points. The method is validated against three-dimensional Laplace and linear elasticity problems, where, in the latter case, displacement and stress fields are expressed via the Papkovich-Neuber potentials. The numerical results show an accurate approximation of both scalar and vector fields, with errors remaining controlled throughout the domain. Overall, the work demonstrates that the incorporation of analytical structures into neural network architectures provides a natural and effective framework for the meshless approximation of three-dimensional boundary value problems while preserving the underlying properties of the governing equations.

The $\ell$-reconfiguration graph of a graph $G$, denoted by $\mathcal{R}_{\ell}(G)$, is the graph whose vertices are the proper $\ell$-colorings of $G$, with an edge between two colorings if they differ in color on exactly one vertex. For any graph $G$ of treewidth at most $2$, Bousquet and Perarnau showed that $\mathcal{R}_\ell(G)$ has linear diameter for $\ell\geq 6$. This result was later extended by Bartier, Bousquet, and Heinrich, who proved that $\mathcal{R}_5(G)$ also has linear diameter. In this paper, we show that for each $\ell\geq 5$, the $\ell$-reconfiguration graphs of $K_{2,3}$-minor-free graphs, some of which include graphs of treewidth $3$, have linear diameter. As a key step in our proof, we also establish that the $(\ell-1)$-reconfiguration graphs of cactus graphs have linear diameter.

We propose a Dynkin-type condition for smooth Riemannian manifolds with boundary. We show that this condition implies bi-Lipschitz equivalence with a Bakry-Émery weighted Riemannian manifold obtained via a time change. As a consequence, we obtain various results, including a local doubling property as well as lower bounds on the Neumann spectral gap and logarithmic Sobolev constant. The local doubling property also yields a new precompactness theorem for manifolds with boundary.

We show that strong log-concavity emerges in probit regression likelihoods without ridge penalization (i.e. Gaussian priors), unlike for the logistic case. Specifically, we provide: (a) a non-asymptotic characterization of strong log-concavity for fixed designs, similar to that for the existence of the maximum likelihood estimator (MLE) and (b) an asymptotic analysis for random designs, in the proportional regime when the sample size $n$ and the number of covariates $d$ grow proportionally $d/n\rightarrow r\in[0,1)$. In the latter case we show that, provided $r$ is small enough, the resulting condition number is finite and bounded independently of $n,d,r$ with high-probability. Numerically tractable estimates are given in case $r=0$. Thus, probit regression provides a non-trivial example of a high-dimensional, well-conditioned log-concave objective.

A linear code is referred to as a locally repairable code (LRC) with locality r if any erased code symbol can be recovered by accessing at most r other code symbols. LRCs are highly desirable for distributed storage systems to enhance repair efficiency. In this paper, we investigate LRCs with disjoint repair sets via the parity-check matrix method. Firstly, we propose a novel concept of the s-Pasch configuration and present a geometric characterization for the existence of LRCs with minimum distance 5 and locality 3. Subsequently, we construct k-optimal LRCs by exploiting the point-line relationship in PG(2,q). Finally, a family of q-ary k-optimal LRCs with minimum distance 6 and general locality r is constructed using partial r-spreads.

Exponential Random Graphs (ERGs) are among the most widely used network models, derived as principled least-bias graph ensembles that maximize Shannon entropy under constraints on the expected values of given structural properties. However, it has been recently (re)discovered that, in the absence of additional information privileging Shannon entropy, the most agnostic inferential construction should maximize the broader class of Uffink entropies. The resulting entropy-maximizing distribution changes from the exponential (Boltzmann-Gibbs) to the so-called q-exponential one. Since maximizing Shannon entropy may produce an unjustified independence between degrees of freedom, here we investigate how the most popular ERGs with independent edges (namely, the Erdos-Renyi and configuration models) generalize to higher-order q-Exponential Random Graphs with dependent edges in the non-Shannon case, while keeping their defining constraints (number of links and degree sequence, respectively) unchanged. We find features, such as a phase transition between sparse and dense regimes, that are absent in the original ERGs but typical of higher-order networks, plus novel phenomena such as richer assortativity and clustering profiles, which allow for the coexistence of link sparsity and triadic closure. These results show that higher-order networks do not necessarily require higher-order constraints, as they naturally arise from simpler ones in a framework that is even more agnostic than Shannon's.

In this paper, we establish general monotone quantities and sharp mass-capacity inequalities related to $p$-capacitary functions in $3$-dimensional asymptotically flat half-spaces of simple topology with nonnegative scalar curvature and nonnegative boundary mean curvature. These inequalities attain equality on a Schwarzschild half-space outside a rotationally symmetric half sphere.

We consider the problem of binary classification in a framework where the predictor $X$ takes values in an arbitrary separable metric space $\mathcal X$ and the label $Y$ values in $\{ \pm 1 \}$. In the first part of this work, we assume that one has direct access to an i.i.d. sample $(X_1,Y_1),\ldots,(X_n,Y_n)$ from the unknown distribution of the pair $(X,Y)$. We derive a convergence rate for the Proto-NN classifier which was recently introduced as a classifier in the presence of metric space-valued predictors. In the second part of the paper, we reconsider the same problem under an additional privacy constraint. More precisely, we work in the framework of local differential privacy where one assumes that the data $(X_1,Y_1),\ldots,(X_n,Y_n)$ cannot be directly observed but only a privatised surrogate obtained through a suitable mechanism satisfying the privacy constraint is available. The statistician should select an optimal privacy mechanism from the class of all mechanism that guarantee local differential privacy. Our method of choice is to add Laplace distributed noise to both a set of in Proto-NN classifier using the privatised data only is universally consistent. Finally, a rate of convergence for the privatised Proto-NN classifier is derived.

We prove an optimal $L^1$ stability theorem for Hölder's inequality. Let $p>1$, $q>1$, and $1/p+1/q=1$. If $a_k,b_k\ge 0$ and \[ \sum_{k=1}^n a_k=\sum_{k=1}^n b_k=1, \] then \[ 1-\sum_{k=1}^n a_k^{1/p}b_k^{1/q} \ge \frac1{2pq}\left(\sum_{k=1}^n |a_k-b_k|\right)^2 . \] The constant $1/(2pq)$ is best possible. We also give the corresponding integral form.

The purpose of this article is to determine the gluing data associated to degeneration of local systems to the boundary of the Baily–Borel compactification of a Shimura variety.

This paper studies sum-of-squares (SoS) representations of nonnegative polynomials over the hypercube $[0,1]^n$. De Klerk and Laurent (SIAM J. Optim., 2010) conjectured that the smallest constant $C_n$ such that the polynomial $x_1\cdots x_n +C_n$ is contained in the degree-$n$ truncated quadratic module $M_{n,n}(x_1-x_1^2,\ldots,x_n-x_n^2)$ of the hypercube is $C_n=1/(n(n+2))$, for $n$ even. We specialize symmetry reduction techniques for finding sum-of-squares certificates to the hypercube, where the generators $x_i-x_i^2$ are not individually invariant under the symmetric group but form an invariant set, and apply them to this conjecture. Combining this reduction with a further (heuristic) sparsity reduction, a rational rounding step, and an exact verification over $\mathbb{Q}$, we prove the bound $C_8\leq 11/1000 <1/80$. In particular, this disproves the conjectured optimal value for $n=8$.

In this thesis we focus on quasilinear elliptic systems driven by various nonlinear operators, such as the p-Laplacian, and nonlinear sources that are allowed to exhibit both subcritical and critical growth. We aim to establish the existence of solutions for perturbation of specific eigenvalue problems, by employing variational and topological methods. To establish existence results for autonomous systems of quasilinear PDEs in the spirit of the paper by Amann and Zehnder, we develop a local Morse theory for functional associated to quasilinear elliptic systems. By refining topological arguments introduced by Cingolani and Degiovanni in Banach product spaces, we establish the finiteness of the critical groups and we derive a Poincaré-Hopf formula in a Banach product space, in presence of both subcritical and critical nonlinear coupling. We also establish uniform boundedness results for anisotropic quasilinear systems, that are of interest within regularity theory. To show existence results for non-autonomous systems of quasilinear PDEs in the spirit of the paper of Landesman and Lazer, we consider the eigenvalue problem for quasilinear elliptic systems introduced by de Thélin. We prove the simplicity and isolation of the first eigenvalue lambda1. Furthermore, we show the existence of a sequence of eigenvalues by employing a suitable deformation lemma proved by Bonnet. Subsequently, we analyze new sufficient Landesman-Lazer type conditions within the framework of quasilinear elliptic systems. We also investigate the N-dimensional Euclidean Onofri inequality, established by Del Pino and Dolbeault for smooth functions with compact support. After extending this inequality to a suitable weighted Sobolev space, we exploit its connection with the Liouville equation on R^N to prove an equivalence with the sharp logarithmic Moser-Trudinger inequality on the unit ball of R^N.

We establish a criterion for the integrality of averages of roots of unity and apply it to settle a conjecture regarding the linearity of functions on $\mathbb{Z}_n$. Specifically, we prove that for any modulus $n \ge 1$, if a function $f: \mathbb{Z}_n \to \mathbb{Z}_n$ satisfies that the averages $\frac{1}{n} \sum_{x=0}^{n-1} \omega^{f(x)+bx}$ (where $\omega=e^{2\pi i/n}$) are algebraic integers for all $b \in \mathbb{Z}_n$, then $f$ is necessarily linear modulo $n$. This provides a short, elementary proof that works uniformly for all $n$ and avoids the finite-field machinery used in previous partial results. Furthermore, when $n=p^r$, we utilize a local-global integrality argument to show that any normalized sum of $p^r$-th roots of unity that is $p$-adically integral must be either $0$ or a single root of unity. As an application, we completely characterize the perfect isometries of the cyclic group $C_{p^r}$: they are precisely those induced by affine permutations $x \mapsto \alpha x + \beta$ with $\gcd(\alpha, p^r)=1$.

This paper studies how efficiently deep ReLU neural networks can approximate and learn smooth functions. When the error is measured in $L^p([0,1]^d)$ norm and the approximator is a network with width $W$ and depth $L$, recent works have proven the supper approximation rate $\mathcal{O}((WL)^{-2s/d})$ for Besov space $\mathcal{B}^s_{q,r}([0,1]^d)$ under the Sobolev embedding condition $s/d>1/q-1/p$. In order to overcome the curse of dimensionality in this rate, we extent this result to anisotropic and mixed smooth function classes. We establish the approximation rate $\mathcal{O}((WL)^{-2\tilde{s}})$ for anisotropic Besov space $\mathcal{B}^{\boldsymbol{s}}_{q,r}([0,1]^d)$ with anisotropic smoothness $\boldsymbol{s}=(s_1,\dots,s_d)$ under the embedding condition $\tilde{s} > 1/q-1/p$, where the mean smoothness $\tilde{s} = (\sum_{i=1}^d s_i^{-1})^{-1}$. For mixed smooth Besov space $\mathcal{MB}^s_{q,r}([0,1]^d)$ with mixed smoothness $s>1/q-1/p$, we show that the approximation rate $\mathcal{O}((WL)^{-2s})$ holds up to logarithmic factors. Using these results, we also derive approximation bounds for the composition of anisotropic Besov functions. As an application, it is shown that deep ReLU neural networks can achieve minimax optimal rates up to logarithmic factors for a wide range of smooth function classes.

Global warming has intensified water scarcity, posing a critical challenge for the agricultural sector, where seasonal demand often leads to significant wastage. This work addresses the need for efficient water management by proposing a smart irrigation control system based on a bilevel optimization algorithm. The framework aims to minimize water consumption across multiple fields simultaneously or, under supply constraints, to distribute available water equitably to minimize deviations from required levels. To achieve this, the problem is modeled hierarchically: the upper level acts as a central controller managing daily water limits and field-wise allocation, while the lower level acts as a centralized scheduler determining the optimal cooperative irrigation execution across all fields. This structure ensures that both global resource constraints and local irrigation needs are met efficiently. Preliminary results from digital simulation tools suggest that the proposed framework significantly improves water-use efficiency and supports sustainable irrigation practices, particularly in water-constrained scenarios.

We describe the group of endotrivial complexes, i.e., the Picard group, of the derived category of permutation modules for a finite group over a commutative Noetherian ring. As a result, we deduce that not every endotrivial with integer coefficients arises from a homotopy representation, i.e., an invertible genuine equivariant spectrum. Along the way, we establish a descent result with respect to subgroups of prime-power order, show that oriented endotrivial complexes are line bundles, that is, locally trivial with respect to an open cover of the Balmer spectrum, and provide a topological construction of forerunner homomorphisms, answering a question of the second-named author.

Nonlinear conservation laws are at the heart of many of the most important dynamical systems in science and engineering. In practical applications, such systems are often subject to various sources of uncertainty, e.g. due to sparse or noisy measurements. Inferring physical quantities and fields of interest then becomes an ill-posed problem which both classical numerical methods and modern deep learning-based methods struggle to treat appropriately. Recent work has framed classical numerical methods as Bayesian inference under Gaussian process priors, resulting in a physics-aware treatment of uncertainties. Following this line of work, we develop a novel numerically conservative method for uncertainty-aware simulations of nonlinear conservation laws. We use recent sparse approximation techniques to scale up to large-scale forward and inverse problems. For forward simulation, we inherit the accuracy of classical solvers while providing structured uncertainty quantification. On inverse problems, we recover posteriors over nonparametric source fields in seconds – outperforming neural baselines that take minutes to produce a less accurate point estimate.

This paper investigates the solution of coupled third-order tensor equation $\mathcal{A} \ltimes \mathcal{X} = \mathcal{B},\ \mathcal{X} \ltimes \mathcal{C} = \mathcal{D},$ of arbitrary dimensions by incorporating semi-tensor product (STP) within t-product framework, where the unknown $\mathcal{X}$ can take form of vector, matrix, or tensor. For the unknown $\mathcal{X}$, we establish a necessary and sufficient condition that provides an equivalence criterion for the existence of solutions. Moreover, the explicit structure (Toeplitz, Circulant) of $\mathcal{C}$ and $\mathcal{D}$ is characterized. Theoretical results are supported by several illustrative examples.

Jacobsthal numbers are an example of a Lucas sequence defined by a recurrence relation analogous to that of the Fibonacci numbers, but with different parameters. In this paper, we prove that the only perfect powers among Jacobsthal numbers are the trivial ones, namely $0$ and $1$. Using the Binet formula, the problem is reduced to an exponential Diophantine equation in three unknowns. We resolve this equation via the modular approach, following the framework developed by Bennett and Skinner for ternary Diophantine equations. This work contributes to the study of perfect powers in linear recurrence sequences.

Gas-particle interactions with non-absorbing surfaces are commonly described using the scattering-kernel formalism. In this framework, an operator $\mathbf{K}$ maps incident velocity distributions to reflected velocity distributions. The operator is self-adjoint and has norm $\lVert \mathbf{K} \rVert = 1$ in an $L^2$ space weighted by the three-dimensional Maxwell-Boltzmann distribution, and must satisfy non-negativity, normalisation, and reciprocity. In standard formulations, $\mathbf{K}$ represents the aggregate effect of all gas-surface interaction mechanisms through a single operator, without distinguishing the physical scales at which these mechanisms occur. For gas scattering from a rough surface, however, it is advantageous to separate geometric effects associated with distinct roughness scales from the underlying thermochemical processes occurring at the atomic scale. We therefore introduce a roughness-based extension of the scattering-kernel formalism, in which a local kernel is successively lifted to larger scales via single- and multi-reflection operators associated with statistically defined surface morphologies. We derive sufficient conditions under which the resulting global kernels preserve reciprocity, normalisation, and non-negativity whenever these properties hold for the smallest-scale kernel. We further show that these constructions define operators on the space of scattering kernels, and establish the associated multi-scale composition laws that allow independent roughness contributions to be combined recursively. The resulting framework provides a general basis for modelling gas-surface scattering on rough surfaces with arbitrary scale decompositions.

For the kinetic Langevin diffusion and its splitting discretizations, the hypoelliptic noise structure makes the relationship between couplings and total variation (TV) bounds more subtle than in the elliptic case. We establish that, for the kinetic Langevin equation with quadratic potential, no Markovian coupling (continuous or discrete) captures the asymptotic decay rate of the TV distance between two solutions with different initial values; the canonical iterated one-shot (or sticky) coupling, for which we derive an exact contraction formula, saturates this lower bound. On the constructive side, we show that the recent sharp TV bounds obtained by Chak and Monmarché admit a natural interpretation through an explicit non-Markovian coupling, built from an optimal coalescence trajectory characterized by a classical minimum-energy control problem. For the OBABO splitting scheme, this approach additionally eliminates the Hessian-Lipschitz, step-size, and final-time assumptions in the work of Chak and Monmarché.

For Z d actions equipped with a completely commuting filtration, normalized partial sums of martingale differences converge in distribution (see [V19]) but even if the action is ergodic, the limit law need not be normal. Explication of this phenomenon has been studied in [GLV] but often in dimension 2 only. It turned out that the non normality is related to the factor of product type I. Here we extend the necessary and sufficient condition of normality from dimension 2 to all finite dimensions, as well as a sufficient condition of normality for martingale differences in the orthocomplement of L 2 (I). At the same time an improved and corrected version of the central limit theorem for random fields from [V19] with dimension d > 2 is given.

We prove operator-norm and gradient Lipschitz bounds for exponential-map parameterizations on the special Euclidean algebra se(3), providing an explicit example of intermediate polynomial growth behavior. Using the contraction property of the SO(3) left Jacobian, we show that ||exp(theta)||_op <= 1 + ||theta||_F for all theta in se(3). We then derive a self-contained O(R^2) upper bound for the gradient Lipschitz constant, with explicit constant 4.02, and construct an objective J* satisfying L_J*(R; se(3)) >= 0.0505 R^2 for R >= 2. These results place se(3) between compact Lie algebras, where the Lipschitz constant remains bounded, and Lie algebras with hyperbolic elements, where it grows exponentially. The upper and lower bounds are obtained for different objective classes; no minimax optimality claim is made.

This paper characterizes the performance limits of optimal array designs using orthogonal and coherent waveforms for both linear and planar arrays. For orthogonal waveforms, we show that the single-target Cramér-Rao Bound (CRB) depends on the sum of the so-called spatial variances of the transmit (Tx) and receive (Rx) arrays, or equivalently, the spatial variance of the sum co-array weighted by the multiplicities of the virtual sensors. This reveals that CRB-optimal geometries are inherently redundant, highlighting a fundamental trade-off between mean squared error (MSE) and identifiability in parameter estimation. Moreover, we derive optimal Tx-Rx sensor allocations given a total sensor budget and show that unequal allocation (favoring the Rx) is optimal even for nonredundant arrays, questioning conventional designs. We extend our results to planar arrays, providing a new general condition that the spatial covariances of the Tx and Rx arrays should satisfy for the optimal waveforms to direct power in the target direction. Additionally, we establish a connection between Diophantine equations and array geometries with equal CRB, along with a constructive method for designing such arrays. Our work provides new guidelines for and insights into optimal array and waveform design with relevance in emerging active sensing multiple-input multiple-output systems.