Peer Revenue Equity in science

Papers

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

Mathematics

math Watch Mathematics (sign in to use)

Thursday 5 March

Combinatorics

Geodesic-transitive graphs with large diameter

We review the nearly complete classification project for finite distance-transitive graphs and compile a list of all known graphs. Interestingly, we find that those graphs with diameter larger than 4, apart from a small finite number of exceptions, are geodesic-transitive. Their geodesics exhibit a clear (often geometric) structure. On the other hand, we provide examples of graphs that are distance-transitive but not geodesic-transitive, including two infinite families with diameter 3 and a few sporadic ones with diameter 3, 4 or 7. In the last section, we extend our investigation to polar Grassmann graphs and provide an explicit description of their geodesics.
Read the full abstract

Analysis of PDEs

Multi-Species Keller–Segel Systems: Analysis, Pattern Formation, and Emerging Mathematical Structures

Chemotaxis systems of Keller–Segel type constitute one of the central mathematical frameworks for understanding aggregation phenomena in biological and ecological systems. Over the past decades, the theory has evolved from the classical single-species model to increasingly sophisticated multi-species and multi-signal formulations that capture competition, cooperation, antagonistic chemotaxis, and interactions with fluid environments. This article provides a comprehensive exposition of multi-species Keller–Segel systems and their mathematical structure. We review fundamental analytical results concerning local and global well-posedness, mechanisms of finite-time blow-up, and the role of critical mass and dimensionality. Particular emphasis is placed on how cross-diffusion, antagonistic interactions, logistic effects, and nonlinear production terms alter the qualitative behavior of solutions. We further examine the mathematical mechanisms underlying pattern formation, including diffusion-driven instabilities, bifurcation phenomena, and the emergence of spatial and spatiotemporal structures. Connections between analytical thresholds and observed nonlinear dynamics are highlighted, and the interplay between reaction kinetics, chemotactic sensitivity, and diffusion is discussed from a unifying perspective. By synthesizing classical results with recent developments, this survey aims to clarify the structural principles governing multi-species chemotaxis systems, identify common analytical techniques, and outline open problems that remain central to the field. The exposition is intended to serve both specialists and researchers entering the area of nonlinear partial differential equations and mathematical biology.
Read the full abstract

All authors Kolade M Owolabi, Eben Mare, Clara O Ijalana, Kolawole S Adegbie

Representation Theory

Horospherical splittings of $\mathfrak g$ and related Poisson commutative subalgebras of $\mathcal S(\mathfrak g)$

Let a Lie algebra $\mathfrak q$ be a linear sum of two complementary subalgebras $\mathfrak h$ and $\mathfrak r$. We continue our investigations initiated in (J. London Math. Soc. 103 (2021), 1577-1595), where compatible Poisson brackets associated with splitting $\mathfrak q=\mathfrak h\oplus\mathfrak r$ and related Poisson-commutative subalgebras of the symmetric algebra $\mathcal S(\mathfrak q)$ are studied. In this article, we further develop the general theory and study in more details splittings of the reductive Lie algebras such that both $\mathfrak h$ and $\mathfrak r$ are solvable horospherical subalgebras. We also derive some results of the Adler-Kostant-Symes theory using our approach.
Read the full abstract

Numerical Analysis

Quantum relative entropy regularization for quantum state tomography

The density matrix is a positive semidefinite operator of trace 1 characterizing the state of a quantum system. We consider the inverse problem to reconstruct such density matrices from indirect measurements, also known as quantum state tomography. To solve such inverse problems in high or infinite dimensional settings, we study variational regularization using the quantum relative entropy as penalty functional. Quantum relative entropy is an analog of the well-known maximum entropy functional with compositions of functions replaced by the spectral functional calculus. The main aim of this paper is to establish the regularizing property of this scheme. As a crucial intermediate step, we establish lower semi-compactness of the penalty functional with respect to the weak-$*$-topology. Moreover, we compute the subgradient, proximal operator, and conjugate functional of the quantum relative entropy on finite dimensional spaces. This enables us to apply iterative algorithms from convex optimization to solve the regularized problems numerically. To show the validity and practical value of our results, we apply our theory to the examples of Photon-Induced Near-field Electron Microscopy (PINEM) and to optical homodyne tomography.
Read the full abstract

Optimization and Control · Cross-listed from cs.RO

U-OBCA: Uncertainty-Aware Optimization-Based Collision Avoidance via Wasserstein Distributionally Robust Chance Constraints

Uncertainties arising from localization error, trajectory prediction errors of the moving obstacles and environmental disturbances pose significant challenges to robot's safe navigation. Existing uncertainty-aware planners often approximate polygon-shaped robots and obstacles using simple geometric primitives such as circles or ellipses. Though computationally convenient, these approximations substantially shrink the feasible space, leading to overly conservative trajectories and even planning failure in narrow environments. In addition, many such methods rely on specific assumptions about noise distributions, which may not hold in practice and thus limit their performance guarantees. To address these limitations, we extend the Optimization-Based Collision Avoidance (OBCA) framework to an uncertainty-aware formulation, termed U-OBCA. The proposed method explicitly accounts for the collision risk between polygon-shaped robots and obstacles by formulating OBCA-based chance constraints, and hence avoiding geometric simplifications and reducing unnecessary conservatism. These probabilistic constraints are further tightened into deterministic nonlinear constraints under mild distributional assumptions, which can be solved efficiently by standard numerical optimization solvers. The proposed approach is validated through theoretical analysis, numerical simulations and real-world experiments. The results demonstrate that U-OBCA significantly mitigates the conservatism in trajectory planning and achieves higher navigation efficiency compared to existing baseline methods, particularly in narrow and cluttered environments.
Read the full abstract

All authors Zehao Wang, Yuxuan Tang, Han Zhang, Jingchuan Wang, Weidong Chen

Optimization and Control

Biquadratic SOS Rank and Augmented Zarankiewicz Number

This paper introduces the concepts of the augmented Zarankiewicz number $z_A(m,n)$ and the limited augmented Zarankiewicz number $z_L(m,n)$, which are natural combinatorial extensions of the classical Zarankiewicz number. These numbers arise from augmented bipartite graphs that may contain both standard edges (1-edges) and pairs of edges representing squares of binomials (2-edges). The main theoretical result establishes the inequality chain $\operatorname{BSR}(m, n) \geq z_A(m, n) \geq z_L(m, n) \geq z(m, n)$, linking the maximum biquadratic sum-of-squares (SOS) rank to these extremal graph parameters. We determine the exact values of $z_L(m, n)$ for the cases $(m,2)$, $(3,3)$, $(4, 3)$ and $(4,4)$, and provide new lower bounds for the cases $(5,3)$, $(5,4)$, and $(5,5)$. These results yield improved lower bounds for the maximum SOS rank of biquadratic forms, demonstrating that $z_L(m,n)$ can exceed the classical Zarankiewicz number, thereby offering a refined combinatorial perspective on the SOS rank problem.
Read the full abstract

Combinatorics

Hat guessing with proper colorings

We initiate the study of the hat guessing number of a graph where the adversary is only allowed to provide a proper coloring of the graph. This is the largest number $q$ for which there is a guessing strategy on each vertex that only depends on its neighborhood, such that for every proper coloring of the graph with $q$ colors at least one vertex guesses its color correctly. In this variation, we prove that the hat guessing number of the complete graphs on $n$ vertices is $2n - 1$, which is roughly twice the classical hat guessing number of the complete graph. Our winning strategy is related to finding perfect matchings between the middle layers of the boolean poset of dimension $2n - 1$. We prove that the hat guessing number of all trees on $n \geq 3$ vertices is equal to $4$. We derive general upper bounds in terms of the number of vertices, chromatic number, and maximum degree, and obtain improved bounds for book graphs. Using our results and an ILP formulation of the problem, we determine the exact hat guessing number for all graphs on at most $4$ vertices, give bounds on graphs on $5$ vertices, and suggest some open problems.
Read the full abstract

All authors Sam Adriaensen, Peter Bentley, Anurag Bishnoi, Wouter Cames van Batenburg, Michael Kreiger, Lars van der Kuil, Saptarshi Mandal, Anurag Ramachandran, James Tuite

Complex Variables

Non-Runge Fatou-Bieberbach Domains in Stein Manifolds with the Density Property and Non-tame Families of Stein Structures

Let $X$ be a Stein manifold with the density property. We present methods of constructing two kinds of non-Runge Fatou-Bieberbach domains in $X$, which by definition are proper open subsets of $X$ biholomorphic either to $\mathbb{C}^n$ with $n = \dim X$ or to $X$. For both kinds, we provide examples where our methods apply. The results are applied to give new examples of non-tame families of Stein structures on manifolds with non-trivial topology.
Read the full abstract

Optimization and Control · Cross-listed from stat.ML

How Does the ReLU Activation Affect the Implicit Bias of Gradient Descent on High-dimensional Neural Network Regression?

Overparameterized ML models, including neural networks, typically induce underdetermined training objectives with multiple global minima. The implicit bias refers to the limiting global minimum that is attained by a common optimization algorithm, such as gradient descent (GD). In this paper, we characterize the implicit bias of GD for training a shallow ReLU model with the squared loss on high-dimensional random features. Prior work (Vardi and Shamir, 2021) showed that the implicit bias does not exist in the worst-case, or corresponds exactly to the minimum-$\ell_2$-norm interpolating solution under exactly orthogonal data (Boursier et al., 2022). Our work interpolates between these two extremes and shows that, for sufficiently high-dimensional random data, the implicit bias approximates the minimum-$\ell_2$-norm solution with high probability with a gap on the order $\Theta(\sqrt{n/||\lambda||_1})$, where $n$ is the number of training examples and $\lambda$ denotes the spectrum of the data covariance matrix. Our results are obtained through a novel primal-dual analysis that carefully tracks the evolution of predictions, data-span coefficients, as well as their interactions, and show that the ReLU activation pattern quickly stabilizes with high probability over random data.
Read the full abstract

All authors Kuo-Wei Lai, Guanghui Wang, Molei Tao, Vidya Muthukumar

Algebraic Geometry

Construction of higher Chow cycles on cyclic coverings of $\mathbb{P}^1 \times \mathbb{P}^1$, Part II

In this paper, we construct higher Chow cycles of type $(2, 1)$ on a family of surfaces related to a product of curves, which are certain degree $N$ abelian covers of $\mathbb{P}^1$ branched over $n+2$ points. We prove that for a very general member, these cycles generate a subgroup of the indecomposable part of $\operatorname{rank} \ge n\cdot \varphi(N)$, where $\varphi(N)$ is Euler's totient function, by computing their images under the transcendental regulator map.
Read the full abstract

Functional Analysis

Global versus regional internal–external potential field separation

Internal–external field separation is crucial for many aspects of geomagnetism, aiming at distinguishing contributions of the magnetic field generated within a given observation surface from those generated in the exterior. When data are available on a full spherical observation surface, this separation is a standard, stable, and widely used procedure dating back to Gauss. However, when data are only available in a subdomain of the observation surface (as is the case for aeromagnetic and ground-based surveys), the situation drastically changes. Here we show that, without prior assumptions, an internal–external field separation is not uniquely possible. Given the geophysically reasonable assumption that the exterior sources, e.g., magnetospheric and ionospheric current systems, are located above a source-free spherical shell, we show that a unique separation becomes possible but that it is highly unstable. The results are based on the spherical Hardy–Hodge decomposition and explain the intrinsic difficulties of regional data-based internal–external potential field separation.
Read the full abstract

Mathematical Physics · Cross-listed from quant-ph

Quantum Weight Reduction with Layer Codes

Quantum weight reduction procedures ease the implementation of quantum codes by sparsifying them, resulting in low-weight checks and low-degree qubits. However, to date, only few quantum weight reduction methods have been explored. In this work we introduce a simple and general procedure for quantum weight reduction that achieves check weight 6 and total qubit degree 6, lower than existing procedures at the cost of a potentially larger qubit overhead. Our quantum weight reduction procedure replaces each qubit and check in an arbitrary Calderbank-Shor-Steane code with an ample patch of surface code, these patches are then joined together to form a geometrically nonlocal Layer Code. This is a quantum analog of the simple classical weight reduction procedure where each bit and check is replaced by a repetition code. Due to the simplicity of our weight reduction procedure, bounds on the weight and degree of the resulting code follow directly from the Layer Code construction and hence are easily verified by inspection. Our procedure is well suited for implementation in modular architectures that consist of surface code patches networked via long-range interconnects.
Read the full abstract

Optimization and Control

A class of stochastic control problems with state constraints

We obtain a probabilistic solution to linear-quadratic optimal control problems with state constraints. Given a closed set $\mathcal{D}\subseteq [0,T]\times\mathbb{R}^d$, a diffusion $X$ in $\mathbb{R}^d$ must be linearly controlled in order to keep the time-space process $(t,X_t)$ inside the set $\mathcal{C}:=([0,T]\times\mathbb{R}^d)\setminus\mathcal{D}$, while at the same time minimising an expected cost that depends on the state $(t,X_t)$ and is quadratic in the speed of the control exerted. We find a probabilistic representation for the value function and an optimal control under a set of mild sufficient conditions concerning the coefficients of the underlying dynamics and the regularity of the set $\mathcal{D}$. The optimally controlled dynamics is in strong form, in the sense that it is adapted to the filtration generated by the driving Brownian motion. Fully explicit formulae are presented in some relevant examples.
Read the full abstract

Analysis of PDEs

The Fractional-Logarithmic Laplacian: Potentials, Regularity, and Critical Compact Embeddings

We develop potential-theoretic and \(L^p\)-regularity results for the fractional–logarithmic Laplacian \((-\Delta)^{s+\ln}\) and its inhomogeneous counterpart \((\lambda I-\Delta)^{s+\ln}\), \(\lambda>1\). These operators lead to logarithmic analogues of the classical Riesz and Bessel potentials. For the associated logarithmic Bessel kernel \(K_{s+\ln}^{\lambda}\), we obtain representation formulas and sharp pointwise asymptotics at both the origin and infinity, including explicit leading constants. A key ingredient is a measure-level bridge between the homogeneous and inhomogeneous symbols. This allows us to pass between the equations $(\lambda I-\Delta)^{s+\ln}u=f$ and $(-\Delta)^{s+\ln}u=f,$ and yields global \(L^p\) estimates, well-posedness for distributional solutions, and a natural scale of logarithmic Bessel spaces \(\mathcal L^p_{s+\ln,\lambda}\). We also discuss the dependence of these spaces on \(\lambda\), their relation to the classical Bessel spaces and with the logarithmic Bessel potential spaces introduced by Opic and Trebels. As applications, we prove endpoint embeddings and critical compactness results. On the critical line \(n=2sp\), we obtain embeddings with a logarithmic modulus of continuity, local compactness on bounded domains, and global compactness in the radial class. In the subcritical case \(n>2sp\), we prove compactness at the critical Sobolev exponent $p^*=\frac{np}{n-2sp},$ recovering compactness at the borderline Lebesgue threshold, a phenomenon absent from the classical Sobolev and Bessel scales.
Read the full abstract

Number Theory

Frequency of a Digit in the Representation of a Number and the Asymptotic Mean Value of the Digits

We study the relationship between the frequency of a ternary digit in a number and the asymptotic mean value of the digits. The conditions for the existence of the asymptotic mean of digits in a ternary number are established. We indicate an infinite everywhere dense set of numbers without frequency of digits but with the asymptotic mean of the digits.
Read the full abstract

Commutative Algebra

Lattice points arising from regularity and $\mathrm{v}$-number of Graphs: Whisker and Cameron-Walker

Let $G$ be a simple graph on $n$ vertices and $I(G)\subseteq R$ be its edge ideal. In this paper, we initiate the study of determining lattice points in $\mathbb{N}^2$ that appear as a pair $(\mathrm{reg}(R/I(G)), \mathrm{v}(I(G)))$, where $G$ ranges over all connected graphs on $n$ vertices, and we denote this set by $\mathcal{RV}(n)$. Here `$\mathrm{reg}$' denotes the (Castelnuovo-Mumford) regularity and `$\mathrm{v}$' denotes the $\mathrm{v}$-number. We establish general bounds for $\mathcal{RV}(n)$ by identifying two sets $A(n)$ and $B(n)$ satisfying $A(n)\subseteq \mathcal{RV}(n)\subseteq B(n)$. Furthermore, we explicitly determine the subsets of $\mathcal{RV}(n)$ consisting of all possible pairs $(\mathrm{reg}(R/I(G)), \mathrm{v}(I(G)))$ arising from whisker graphs and Cameron-Walker graphs on $n$ vertices. Finally, we propose a conjecture on the subset of $\mathcal{RV}(n)$ arising from connected chordal graphs.
Read the full abstract

Number Theory

Transformations and functions that preserve the asymptotic mean of digits in the ternary representation of a number

In the paper we study transformations of the interval $[0;1)$ and functions that preserve the asymptotic mean $r$ of the digits in the $s$–adic representation of a number $x$, $$r(x)=\lim\limits_{n\to\infty}\frac{1}{n}\sum\limits^{n}_{i=1}\alpha_i(x)$$ specify the necessary and sufficient conditions for a transformation to belong to this class
Read the full abstract

Information Theory · Cross-listed from cs.CE

Set-Membership Localization via Range Measurements

In this paper we discuss a classical geometrical problem of estimating an unknown point's location in $\Real{n}$ from several noisy measurements of the Euclidean distances from this point to a set of known reference points (anchors). We approach the problem via a set-mem\-ber\-ship methodology, in which we assume the distance measurements to be affected by unknown-but-bounded errors, and we characterize the set of all points that are consistent with the measurements and their assumed error model. This set is nonconvex, but we show in the paper that it is contained in a region given by the intersection of certain closed balls and a polytope, which we call the localization set. Then, we develop efficient methods, based on convex programming, for computing a tight outer-bounding set of simple structure (a box, or an ellipsoid) for the localization set, which then acts as a guaranteed set-valued location estimate. % The center of the bounding set also serves as a point location estimate. Related problems of inner approximation of the localization set via balls and ellipsoids are also posed as convex programming problems. Different from existing methods based on semidefinite programming relaxations of a nonconvex cost minimization problem, our approach is direct, geometric and based on a polyhedral set of points that satisfy pairwise differences of the measurement equations.
Read the full abstract

Algebraic Geometry

Stability conditions on noncommutative crepant resolutions of 3-dimensional isolated singularities

Let $R$ be a 3-dimensional complete local Gorenstein isolated singularity. For a basic maximal modifying $R$-module $M$, we construct a wall-and-chamber structure, denoted by ${\sf Cone}(M)$ and called the mutation cone of $M$, in the real Grothendieck group associated to the maximal modification algebra $\Lambda={\rm End}_R(M)$. Each chamber in ${\sf Cone}(M)$ corresponds to a maximal modifying module obtained by iterated (Iyama–Wemyss) mutations of $M$, and a wall-crossing corresponds to the mutation at an indecomposable summand. Moreover, we introduce the notion of tilting-noetherian property of $\Lambda$, and by analysis of wall-and-chamber structure of ${\sf Cone}(M)$, we prove that this property holds for $\Lambda$ if and only if all maximal modifying $R$-modules are connected by iterated mutations. We then consider the finite length subcategory $\mathscr{D}_M\subset {\rm D}^{\rm b}({\rm mod}\,\Lambda)$ and introduce a full-dimensional connected subspace ${\rm Stab}^{\rm mdf}\mathscr{D}_M\subset{\rm Stab}\mathscr{D}_M$ of Bridgeland stability conditions on $\mathscr{D}_M$. We prove that there is a regular covering map from ${\rm Stab}^{\rm mdf}\mathscr{D}_M$ to the complexification ${\sf Cone}(M)_{\mathbb{C}}$ of the mutation cone of $M$, where the Galois group is the subgroup of ${\rm Auteq} \mathscr{D}_M$ consisting of compositions of equivalences associated to mutations of maximal modifying modules. Finally, using the results on stability conditions, we describe the group of autoequivalences of $\mathscr{D}_M$ that preserve the subspace ${\rm Stab}^{\rm mdf}\mathscr{D}_M$.
Read the full abstract

2603.04858 · 5 Mar · A crucial error was found in the assertion of Lemma 3.12 (2), which affects the validity of the main results. Although the results are still valid in the case of a terminal singularity, many results for such a case are already known. Therefore we have decided to withdraw the paper

Category Theory

Generalized Gorenstein Categories

Let $\mathscr{A}$ be an abelian category and let $\mathscr{C}$ and $\mathscr{D}$ be additive subcategories of $\mathscr{A}$. As a generalization of Gorenstein categories, we introduce one-sided $n$-$(\C,\D)$-Gorenstein categories with $n\geq 0$. Under certain conditions, we give some equivalent characterizations of one-sided $n$-$(\C,\D)$-Gorenstein categories in term of the finiteness of projective and injective dimensions relative to one-sided Gorenstein subcategories, which induce some new equivalent characterizations of Gorenstein categories. Then we apply these results to categories of interest. In particular, a necessary condition is obtained for the validity of the Wakamatsu tilting conjecture.
Read the full abstract

Optimization and Control

Policy Optimization of Mixed H2/H-infinity Control: Benign Nonconvexity and Global Optimality

Mixed H2/H-infinity control balances performance and robustness by minimizing an H2 cost bound subject to an H-infinity constraint. However, classical Riccati/LMI solutions offer limited insight into the nonconvex optimization landscape and do not readily scale to large-scale or data-driven settings. In this paper, we revisit mixed H2/H-infinity control from a modern policy optimization viewpoint, including the general two-channel and single-channel cases. One central result is that both cases enjoy a benign nonconvex structure: every stationary point is globally optimal. We characterize the H-infinity-constrained feasible set, which is open, path-connected, with boundary given exactly by policies saturating the H-infinity constraint. We also show that the mixed objective is real analytic in the interior with explicit gradient formulas. Our key analysis builds on an Extended Convex Lifting (ECL) framework that bridges nonconvex policy optimization and convex reformulations. The ECL constructions rely on non-strict Riccati inequalities that allow us to characterize global optimality. These insights reveal hidden convexity in mixed H2/H-infinity control and facilitate the design of scalable policy iteration methods in large-scale settings.
Read the full abstract

All authors Chih-Fan Pai, Yuto Watanabe, Yujie Tang, Yang Zheng

Differential Geometry

Bergman kernels and Poincaré series

We show that the Bergman kernel of a finite-volume quotient of a Hermitian manifold $\widetilde{X}$ with bounded geometry by a discrete group $\Gamma$ of its isometries is the same as the averaging over $\Gamma$ of the Bergman kernel on $\widetilde{X}$. We then use these results when $\widetilde{X}$ is a Hermitian symmetric space to show that a large class of relative Poincaré series does not vanish. This extends the results of Borthwick-Paul-Uribe and Barron (formerly Foth) to the case of general locally symmetric spaces of finite volume.
Read the full abstract

All authors Louis Ioos, Wen Lu, Xiaonan Ma, George Marinescu

K-Theory and Homology

The Hochschild cohomlogy ring of a self-injective Nakayama algebra is a Batalin-Vilkovisky algebra

Lambre, Zhou and Zimmermann showed that the Hochschild cohomology ring of a Frobenius algebra with semisimple Nakayama automorphism is a Batalin-Vilkovisky algebra. They asked whether the semisimplicity condition is necessary. In this paper, we show that for a self-injective Nakayama algebra, the Hochschild cohomology ring is always a Batalin-Vilkovisky algebra. In course of proofs, we correct some inaccuracies in the literature, hoping not to introduce new errors.
Read the full abstract

All authors Xiuli Bian, Tomohiro Itagaki, Wen Kou, Weiguo Lyu, Guodong Zhou

Probability

BBP Phase Transition for a Doubly Sparse Deformed Model

We prove the equivalent of the Baik, Ben Arous, Péché (2004) phenomenon for a novel, doubly sparse model where both the Wigner noise matrix and signal vector(s) are sparse. Specifically, we consider a deformed sub-Gaussian sparse Wigner ensemble with a fixed number of sub-Gaussian spike vectors of the same-order sparsity added. We show that spike vectors with signals greater than one are correlated with the top eigenvectors of the deformed ensemble and that each spike vector of signal greater than one induces an outlier eigenvalue. Notably, our results hold in the supercritical sparsity regime for the Wigner matrix ($q \gg \frac{\log n}{n}$) and for any sparse spike vector with an unbounded number of entries ($np\to \infty$). No further relationship between the sparsities of the noise matrix ($q$) and spike vectors ($p$) is necessary. This generalizes the work of Benaych-Georges and Nadakuditi (2010) and Péché (2005).
Read the full abstract

Numerical Analysis · Cross-listed from cs.LG

Multilevel Training for Kolmogorov Arnold Networks

Algorithmic speedup of training common neural architectures is made difficult by the lack of structure guaranteed by the function compositions inherent to such networks. In contrast to multilayer perceptrons (MLPs), Kolmogorov-Arnold networks (KANs) provide more structure by expanding learned activations in a specified basis. This paper exploits this structure to develop practical algorithms and theoretical insights, yielding training speedup via multilevel training for KANs. To do so, we first establish an equivalence between KANs with spline basis functions and multichannel MLPs with power ReLU activations through a linear change of basis. We then analyze how this change of basis affects the geometry of gradient-based optimization with respect to spline knots. The KANs change-of-basis motivates a multilevel training approach, where we train a sequence of KANs naturally defined through a uniform refinement of spline knots with analytic geometric interpolation operators between models. The interpolation scheme enables a “properly nested hierarchy” of architectures, ensuring that interpolation to a fine model preserves the progress made on coarse models, while the compact support of spline basis functions ensures complementary optimization on subsequent levels. Numerical experiments demonstrate that our multilevel training approach can achieve orders of magnitude improvement in accuracy over conventional methods to train comparable KANs or MLPs, particularly for physics informed neural networks. Finally, this work demonstrates how principled design of neural networks can lead to exploitable structure, and in this case, multilevel algorithms that can dramatically improve training performance.
Read the full abstract

All authors Ben S. Southworth, Jonas A. Actor, Graham Harper, Eric C. Cyr

Differential Geometry

The Archimedean height pairing for differential forms on degeneration of Riemann surfaces

We define the Archimedean height pairing for fiberwise cohomologically trivial differential forms on a one-parameter degeneration of Riemann surfaces, and we study its asymptotic behavior. The proof relies on recent work by Dai–Yoshikawa on the asymptotics of small eigenvalues. As an application, we relate this pairing to the current-valued pairing of Filip–Tosatti, extending their construction to broader geometric settings.
Read the full abstract

Logic

The Conjugacy Relation on One-sided Subshifts is Non-treeable

In this paper we study the conjugacy relation on one-sided subshifts in the viewpoint of descriptive set theory. We show the conjugacy relation on one sided subshifts with the alphabet set $\{0,1\}$ is non-treeable and non-amenable.

Combinatorics

The minimum length of an axis-aligned rectangular tiling of a flat torus

A flat torus is the quotient of the Euclidean plane over a lattice generated by a basis, and an axis-aligned rectangular tiling of a flat torus is a partition into finitely many rectangles whose sides are axis-aligned. We provide the minimum sum of the perimeter of rectangles for an axis-aligned rectangular tiling, and prove that it is attainable by either exactly one rectangle or exactly two rectangles.
Read the full abstract

Statistics Theory · Cross-listed from stat.ME

Robust estimation via $\gamma$-divergence for diffusion processes

This paper deals with the problem of outliers in high frequency observation data from diffusion processes. Robust estimation methods are needed because the inclusion of outliers can lead to incorrect statistical inference even in the diffusion process. To construct a robust estimator, we first approximate the transition density of the diffusion process to the Gaussian density by using Kessler's approach and then employ two types of minimum robust divergence estimation methods. In this paper, we provide the asymptotic properties of the robust estimator using $\gamma$-divergence. Furthermore, we derive the conditional influence functions of the estimation using divergences and discuss its boundness.
Read the full abstract

Functional Analysis

Cotype of random polytopes

For $N\geq n$, let $P_{N,n}$ be a random polytope in ${\mathbb R}^n$ with vertices $\pm X_i$, $1\leq i\leq N$, where $X_1,\dots,X_N$ are i.i.d standard Gaussian vectors in ${\mathbb R}^n$. Random polytopes $P_{N,n}$, as well as their duals, are classical objects of interest in high-dimensional convex geometry and local Banach space theory. In this paper, we provide a dimension-independent bound on the cotype of the corresponding normed space $({\mathbb R}^n,\|\cdot\|_{P_{N,n}})$, generated by $P_{N,n}$. Let $K'\geq K>1$, and assume that $K'\geq \frac{N}{n}\geq K$. We show that with probability $1-o(1)$, for any $k\geq 1$, and any collection $y_1,\dots,y_k$ of vectors in ${\mathbb R}^n$, $$ {\mathbb E}_\sigma\,\Big\|\sum_{i=1}^k \sigma_i y_i\Big\|_{P_{N,n}}^q \geq \frac{1}{C_q^q}\sum_{i=1}^k \big\|y_i\big\|_{P_{N,n}}^q, $$ where $\sigma=(\sigma_1,\dots,\sigma_k)$ is a vector of random signs, and where $q\in [2,\infty)$ and $C_q\in[1,\infty)$ may only depend on $K,K'$. We discuss the result in context of infinite-dimensional Banach spaces.
Read the full abstract