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)

Tuesday 14 April

Probability · Cross-listed from stat.AP

Cyclic fractional Gaussian noise: time and frequency domain properties

This article introduces cyclic fractional Gaussian noise (cfGn), a stochastic model that integrates second-order cyclostationarity with long-range dependence property. While classical cyclostationary processes are widely discussed in the literature, they often lack the capacity to account for the persistent, slow-decaying correlations found in complex empirical data. To bridge this gap, we extend the amplitude-modulated stationary framework by utilizing increments of two-dimensional fractional Brownian motion (2d fBm) as the underlying driving process. The proposed cfGn model is constructed by summing two components, which include periodic deterministic functions modulating the univariate coordinates of 2d fGn. We provide a rigorous derivation of the considered model's properties, specifically the autocovariance function (ACVF) and frequency-domain characteristics, including the cyclic spectrum. Through theoretical considerations of asymptotic properties and Monte Carlo simulations, we demonstrate that cfGn preserves periodic behavior of ACVF while inheriting long-memory traits which is manifested in time and frequency domains. This framework offers a robust foundation, for instance, in signal-based condition monitoring in systems where periodic fault components coexist with long-range dependent background noise.
Read the full abstract

Optimization and Control

Sliced Wasserstein Steering between Gaussian Measures

Optimal transport with quadratic cost provides a geometric framework for steering an ensemble, modeled by a probability law, with minimal effort. Yet ambient-space formulations become unwieldy in high dimensions, and sensing or actuation in practice often reveals only linear views of the state – camera silhouettes, LiDAR beams, tomographic slices. We develop a sliced feedback controller for distribution steering: the evolving law is projected onto one-dimensional directions on the sphere, the optimal one-dimensional velocity is synthesized in each projection, and these velocities are averaged to produce a feedback control in the ambient space. The construction reduces to the Benamou–Brenier problem in one dimension. In addition, it is invariant under orthogonal transforms, nonexpansive under projections, and well posed on $\mathcal{P}_2(\mathbb{R}^n)$. Computation proceeds by sampling directions on the sphere and solving independent one-dimensional subproblems, yielding a scalable method aligned with partial observations. In the Gaussian setting, we show that the developed sliced controller steers the law to the prescribed target. Furthermore, we derive an identity relating the energy consumption incurred by the controller to the sliced Wasserstein distance.
Read the full abstract

Information Theory

Anchor-Aided Multi-User Semantic Communication with Adaptive Decoders

Semantic communication (SemCom) is accelerating its momentum to catch up with the massive increase in users' demands in both quantity and quality, with the assistance of advanced deep learning (DL) techniques. Specifically, SemCom can actively embed the semantic meaning of the data into the transmission process, while eliminating statistical redundancy to preserve bandwidth resources for other users. Therefore, the transmitter encodes the message in the most concise way, while the receiver tries to interpret the message with the DL model and its knowledge of the transmitter's intended meaning. Most existing works only consider one transmitter and one receiver, which limits their ability to address the diversity in users' models and capabilities. Therefore, in this paper, we propose a multi-user semantic communication system where each user is equipped with a distinct DL-based joint source-channel decoder architecture, reflecting the diversity in computing capacity. The challenging issue with the proposed system is the catastrophic forgetting property of neural networks, where the DL-based encoder fails to encode the data for the previous user when being trained with a new user. To address this, we propose an anchor decoder with an architecture that is symmetric to the encoder. The symmetric decoder has the same computational capacity as the encoder, providing feedback that aligns with the encoder's extraction capabilities and enhances optimization efficiency. The parameters of the optimized encoder are then frozen and used to train decoders for various users, aligning them with the encoder outputs. Finally, we conduct a series of simulation experiments to validate the proposed framework against other benchmarks.
Read the full abstract

All authors Loc X. Nguyen, Phuong-Nam Tran, Trung Thanh Pham, Avi Deb Raha, Eui-Nam Huh, Zhu Han, Choong Seon Hong

Mathematical Physics

Metric-Deformed Heisenberg Algebras and the $q$-Dirac Operator

We introduce a family of metric-deformed Heisenberg algebras $M_1$ and $M_2$, where the commutation relations are expressed directly in terms of the components of a diagonal Lorentzian metric. We show that these algebras unify several known $q$-deformed Heisenberg algebras, including the $q$-$\hbar$ algebra, the new $q$-Heisenberg algebra, and the $q$-generalized Heisenberg algebra, which embed as special cases. Using Sylvester's theorem of inertia, we establish a connection between the metric signature and the deformation parameters. We construct a $q$-Dirac operator $D_q$ from the deformed D'Alembertian and prove that $D_q^2$ recovers the deformed Klein-Gordon operator. Furthermore, we relate this construction to the quadratic $q$-Dirac operator previously introduced by the author, providing a unified framework that bridges spacetime geometry and $q$-deformed quantum algebras.
Read the full abstract

Mathematical Physics · Cross-listed from hep-th

Quasi-Local Celestial Charges and Multipoles

We extend Penrose's quasi-local mass definition to include higher-spin charges associated with the celestial $Lw_{1+\infty}$ symmetries and relate them to traditional definitions of multipoles. The resulting formulae provide explicit expressions that can be computed on finite 2-surfaces, given a choice of null hypersurface. They yield a geometric definition of celestial symmetries and multipoles in generic spacetimes in terms of higher-valence solutions to the twistor equations. This, in turn, gives rise to natural flux-balance laws along the null hypersurface. We also present a first-principles phase-space derivation of these charges, starting from a twistor space action for self-dual gravity that can be identified with the standard gravitational asymptotic phase space at null infinity, performing a large gauge transformation analysis and using the Penrose transform to connect with the corresponding spacetime expressions. Finally, we formulate the spacetime analysis in the Plebanski gauge and relate the celestial symmetries to the integrability of self-dual gravity in the case of a self-dual background.
Read the full abstract

Optimization and Control

Network Epidemic Control via Model Predictive Control: Extended Version

Balancing the societal costs of non-pharmaceutical interventions with epidemic suppression requires adaptive feedback control. Rather than relying on state-dependent operational caps, we formulate an infinite-horizon optimal control problem for a networked SIQR model that strictly enforces suppression via a hard spectral constraint on the transmission dynamics. We derive a safety-critical Model Predictive Control (MPC) approximation that embeds this spectral certificate stage-wise, yielding a tunable exponential decay rate. Furthermore, we construct a terminal set ensuring recursive feasibility and a feasible continuation that decays globally, proving positive invariance directly via the physical depletion of susceptibles rather than standard quadratic Lyapunov functions. To handle prediction uncertainty, we develop a robust counterpart that replaces nominal constraints by upper-envelope versions, recovering recursive feasibility and finite-horizon realized decay. We conclude by validating our approaches using simulation studies that leverage public data from counties in the state of Massachusetts.
Read the full abstract

All authors Mahtab Talaei, Alex Olshevsky, Laura F. White, Ioannis Ch. Paschalidis

Differential Geometry

Spectral Selection and Minimal Morse Structures on the Poincaré Dodecahedral Space

We study the long time behavior of the heat equation on the spherical Poincare dodecahedral space and introduce a spectral selection property P, asserting that for a dense open set of initial data, the solution eventually becomes a minimal Morse function. We first establish an obstruction principle. If the first positive eigenspace of the Laplace Beltrami operator contains a Morse function that is not minimal, then property P fails. Using an explicit representation theoretic description of the spherical first eigenspace, we show that the round metric on M violates property P. We then develop a perturbative spectral selection mechanism. Using conformal variations and a finite dimensional reduction of the first-order splitting of the lowest eigenvalue cluster, we construct metrics arbitrarily close to the spherical metric for which the first eigenvalue is simple and the corresponding eigenfunction is minimal Morse with exactly six critical points. As a consequence, these nearby metrics satisfy property P. This establishes both the failure and the restoration of minimal Morse selection on M, and provides a concrete spectral mechanism linking representation theory, eigenvalue splitting, and global Morse structure.
Read the full abstract

Spectral Theory

Magnetic Dirichlet Laplacian on deformed waveguides

It is well known that the spectrum of the Dirichlet Laplacian for a two-dimensional waveguide, which is a local deformation of a straight strip, is unstable with respect to waveguide boundary deformations. This means that, when the waveguide is a straight strip, the spectrum of the Dirichlet Laplacian is purely essential. On the other hand, local boundary perturbations of the straight strip produce eigenvalues below the essential spectrum. This paper considers the Dirichlet-Laplace operator with a compactly supported magnetic field. Furthermore, we omit the condition that the boundary perturbation is local. We prove that, in this case, the spectrum of the magnetic Laplacian is stable under small deformations of the waveguide boundary.
Read the full abstract

Analysis of PDEs

The Ladyzhenskaya-Prodi-Serrin Conditions and the Search for Extreme Behavior in 3D Navier-Stokes Flows

In this investigation, we conduct a systematic computational search for potential singularities in 3D Navier-Stokes flows on a periodic domain $\Omega$ based on the Ladyzhenskaya-Prodi-Serrin conditions. They assert that for a solution $\mathbf{u}(t)$ of the Navier-Stokes system to be regular on an interval $[0,T]$, the integral $\int_{0}^T \|\mathbf{u}(t)\|_{L^q}^p\,dt$, where $2/p+3/q=1,\;q>3$, and the expression $\sup_{t \in [0,T]} \|\mathbf{u}(t)\|_{L^3}$ must be bounded. Flows which might become singular and violate these conditions are sought by solving a family of variational PDE optimization problems where we identify initial conditions $\mathbf{u}_{0}$ with the corresponding flows $\mathbf{u}(t)$ locally maximizing the integral $\int_{0}^T \|\mathbf{u}(t)\|_{L^q}^p\,dt$ for a range of different values of $q$ and $p$ or the norm $\|\mathbf{u}(T)\|_{L^3}$ for different time windows $T$ and increasing sizes $\| \mathbf{u}_0 \|_{L^q}$ of the initial data. We consider two formulations where these expressions are maximized over appropriate Lebesgue spaces $L^q(\Omega)$ or the largest Hilbert-Sobolev spaces $H^s(\Omega)$ embedded in them. The lack of Hilbert-space structure in the first case necessitates development of a novel computational approach to solve the problem. While no evidence of unbounded growth of the quantities of interest, and hence also for singularity formation, was detected, we were able to quantify how "close" the flows realizing such worst-case scenarios come to forming a singularity. A comparison of these results with estimates on the rate of growth of the norms $||\mathbf{u}(t)||_{L^q}$ and of the enstrophy $\mathcal{E}(t)$ indicates that the extreme flows do enter a regime where these quantities are amplified at a rate consistent with singularity formation in finite time, but this growth is not sustained long enough for singularities to form.
Read the full abstract

Analysis of PDEs

Derivation of effective kinetic equations describing oscillations in viscoelasticity and in compressible Navier-Stokes

These lecture notes are devoted to solutions of hyperbolic-parabolic systems with persistent oscillations. We consider two examples both from mechanics: (i) The system of viscoelasticity of Kelvin-Voigt type with strain energies involving double well potentials, as employed in phase transitions. (ii) The compressible Navier-Stokes equations for a barotropic gas. For each system we construct solutions with persistent oscillations. In a later part we consider the nonlinear homogenization problem. For the systems of viscoelasticity in one-space dimension in Lagrangian coordinates, and for the compressible Navier-Stokes system for barotropic fluids we show how ideas from the kinetic formulation of conservation laws can be used to derive effective equations. The effective equation consists by a kinetic equation coupled with the macroscopic flow.
Read the full abstract

Analysis of PDEs

Dimension Bound of Singular Set of One-Phase Free Boundary Problems in Spaces with Two-Sided Ricci Bound

In this article, we show that for one-phase free boundary problems in noncollapsed limits of $n$-dimensional manifolds with two-sided Ricci curvature bounds, the Hausdorff dimension of the singular set of the free boundary can be bounded by $n-5$, which is sharp in this context.

General Topology

Topological Complexity and Phase Space Stability: A Persistent Homology Approach to Cryptocurrency Risk

Traditional risk measures in finance, predominantly based on the second moment of return distributions or tail risk heuristics (VaR/CVaR), fail to account for the intrinsic geometric structure of market dynamics. This paper introduces a rigorous mathematical framework utilizing Topological Data Analysis (TDA) to quantify risk as the structural instability of the reconstructed phase space. By applying Takens' Delay Embedding Theorem to cryptocurrency log-returns, we generate a point cloud representation of the underlying attractor. We analyze the evolution of the filtration of Vietoris-Rips complexes to compute persistent homology groups $H_k$. We define a "Topological Persistence Norm" to characterize market regimes and propose a leverage calibration heuristic based on the persistence of 1-dimensional cycles. This approach provides a coordinate-free, stability-invariant metric for risk assessment that is robust to high-frequency noise.
Read the full abstract

Functional Analysis

Higher-Order Autocorrelations on Finite Abelian Groups

The question of determining a signal from its higher-order autocorrelation data is of practical interest in fields as varied as X-ray crystallography, image processing, and satellite communications. At the heart of the issue is how much of this autocorrelation data one truly needs. We prove two new upper bounds on the order of data needed to determine a signal on a general (i.e. not necessarily cyclic) finite abelian group depending on some knowledge of the vanishing of the signal's Fourier transform. In investigating lower bounds on the required data, we classify signals on $\mathbb{Z}_6$ not determined by their fifth-order data and provide analogous examples on $\mathbb{Z}_{30}$.
Read the full abstract

Analysis of PDEs

Harnack inequality for non-uniformly elliptic equations in non-divergence form

We study regularity properties for solutions to the nakedly degenerate elliptic equation $a_{ij}\partial_{ij}u =0$, where the coefficients satisfy $I \ge a_{ij}(x) \ge \lambda(x) I$ and the only assumption is that $\lambda^{-1} \in L^p$. We prove an improvement of oscillation and a Liouville theorem for $p>d-1$, and a Harnack inequality for $p$ sufficiently large depending on dimension. Along the way, we obtain a new $\log-L^\varepsilon$ Weak Harnack inequality for supersolutions. Then, touching subsolutions by double exponential blow-up barriers, we also derive a logarithmic local maximum principle that is new even in the uniformly elliptic case. Both of these results hold for $p>d-1$. Finally, we construct examples showing that there cannot be Harnack or Weak Harnack inequalities in the regime $p<d-1$, nor can there be power-type $L^\varepsilon$ inequalities in the case of any $p<\infty$.
Read the full abstract

Optimization and Control

A new perspective on the Løkka-Zervos dichotomy for absolutely continuous dividend strategies

We revisit the optimization problem (and its dichotomous solution) analyzed in Renaud et al. (2026). By choosing affine functions (of the surplus level) to bound the dividend rates, we are able to provide a more explicit derivation of the solution, avoiding the use of viscosity solutions. Moreover, with explicitly parameterized expressions, we are able to study analytical properties of the optimal thresholds and to refine the statement of the dichotomy. To reach these objectives, we add a penalty-at-ruin parameter in the performance function, allowing us to unify the expressions of the value functions of the two subproblems appearing in the dichotomous solution; as a by-product, the dichotomy can also be expressed in terms of this newly added parameter. Finally, we perform sensitivity analyses to illustrate that the range of values generated in an affine-bound framework is flexible and can span the range from a constant bound to the singular version of this problem.
Read the full abstract

All authors Tommy Mastromonaco, Nacer Fendri, Jean-François Renaud, Clarence Simard

Analysis of PDEs

Refined temporal asymptotics near blow-up points in the planar Keller-Segel system

For the Keller-Segel system \[ \left\{\, \begin{aligned} u_t &= \Delta u - \nabla \cdot ( u \nabla v ), \\ v_t &= \Delta v - v + u \end{aligned} \right. \tag{$\star$} \] posed in a planar domain $\Omega$ with Neumann boundary conditions, the existence of classical solutions blowing up at some finite time $T$ has long been established. In fact, it has been shown that for every blow-up point $x$ the quantity $\int_{B_R(x)\cap\Omega} u(\cdot,t )\ln(u(\cdot, t))$ is unbounded as $t\nearrow T$ for all $R > 0$ even though the global mass of $u$ is always conserved. The present manuscript provides some quantitative information on the behavior of such localized $L\log L$ expressions by asserting the existence of $\delta_0=\delta_0(\Omega)>0$ such that any solution to the Neumann problem for ($\star$) blowing up at time $T\in (0,\infty)$ satisfies \[ \limsup_{t\nearrow T} \frac{1}{\ln\frac{T}{T-t}}\int_{B_R(x)\cap\Omega} u(\cdot, t)\ln(u(\cdot, t)) \ge \delta_0 \tag{$\star\star$} \] for all $R > 0$ at each blow-up point $x$. This confirms a certain universality property of the blow-up mechanism seen in the particular examples of radial collapsing solutions constructed in the seminal work [16], especially also beyond the realm of symmetry; apart from that, along with a consequence of ($\star\star$) on the corresponding asymptotics of similarly localized $L^p$ norms of $u$ for $p\in (1,\infty]$, this provides some extension of a known result on non-degeneracy of blow-up points that has concentrated on the choice $p=\infty$ here.
Read the full abstract

Dynamical Systems · Cross-listed from eess.SY

Structure- and Stability-Preserving Learning of Port-Hamiltonian Systems

This paper investigates the problem of data-driven modeling of port-Hamiltonian systems while preserving their intrinsic Hamiltonian structure and stability properties. We propose a novel neural-network-based port-Hamiltonian modeling technique that relaxes the convexity constraint commonly imposed by neural network-based Hamiltonian approximations, thereby improving the expressiveness and generalization capability of the model. By removing this restriction, the proposed approach enables the use of more general non-convex Hamiltonian representations to enhance modeling flexibility and accuracy. Furthermore, the proposed method incorporates information about stable equilibria into the learning process, allowing the learned model to preserve the stability of multiple isolated equilibria rather than being restricted to a single equilibrium as in conventional methods. Two numerical experiments are conducted to validate the effectiveness of the proposed approach and demonstrate its ability to achieve more accurate structure- and stability-preserving learning of port-Hamiltonian systems compared with a baseline method.
Read the full abstract

Probability · Cross-listed from cs.LG

Some Theoretical Limitations of t-SNE

t-SNE has gained popularity as a dimension reduction technique, especially for visualizing data. It is well-known that all dimension reduction techniques may lose important features of the data. We provide a mathematical framework for understanding this loss for t-SNE by establishing a number of results in different scenarios showing how important features of data are lost by using t-SNE.
Read the full abstract

Statistics Theory · Cross-listed from stat.ME

Addressing Confounding by Indication Through (Un)Measured Centre Characteristics in Learn-As-you-GO(LAGO) Trials

The Learn-As-you-Go (LAGO) design is an adaptive clinical trial design allowing modifications to multicomponent intervention packages across stages. Centers participate in more than one stage, as is common in large-scale implementation trials. In LAGO trials, center characteristics may act as confounders, predicting both the intervention package and the outcomes. We extend LAGO theory by introducing fixed center effects to control for confounding by indication through measured and unmeasured center characteristics. Conditioning on center characteristics by including fixed center effects ensures asymptotic results hold without requiring explicit characterization of unmeasured confounders. Our methods apply even with small numbers of centers. LAGO theory is established for continuous outcomes following a generalized linear model and binary outcomes following a logistic regression model, unifying theory across outcome types. Point- and interval estimators are derived, and consistency and asymptotic normality are established. Valid hypothesis tests for the overall intervention effect are provided, and the optimal intervention package minimizing cost subject to a target outcome mean is obtained via constrained optimization.
Read the full abstract

All authors Minh Thu Bui, Christopher T. Longenecker, Ante Bing, Donna Spiegelman, Allison R. Webel, Hayden B. Bosworth, Judith J. Lok

Statistics Theory

Sequential Change Detection for Multiple Data Streams with Differential Privacy

Sequential change-point detection seeks to rapidly identify distributional changes in streaming data while controlling false alarms. Existing multi-stream detection methods typically rely on non-private access to raw observations or intermediate statistics, limiting their usage in privacy-sensitive settings. We study sequential change-point detection for multiple data streams under differential privacy constraints. We consider multiple independent streams undergoing a synchronized change at an unknown time and in an unknown subset of streams, and propose DP-SUM-CUSUM, a differentially private detection procedure based on the summation of per-stream CUSUM statistics with calibrated Laplace noise injection. We show that DP-SUM-CUSUM satisfies sequential $\varepsilon$-differential privacy and derive bounds on the average run length to false alarm and the worst-case average detection delay, explicitly characterizing the privacy–efficiency tradeoff. A truncation-based extension is also presented to handle distributional shifts with unbounded log-likelihood ratios. Simulations and experiments on an Internet of Things (IoT) botnet dataset validate the proposed approach.
Read the full abstract

Optimization and Control

A Momentum-based Stochastic Algorithm for Linearly Constrained Nonconvex Optimization

This paper studies a stochastic algorithm for linearly constrained nonconvex optimization, where the objective function is smooth but only unbiased stochastic gradients with bounded variance are available. We propose a momentum-based augmented Lagrangian method that employs a Polyak-type gradient estimator and requires only one stochastic gradient evaluation per iteration. Under the standard stochastic oracle model and the smoothness condition of the expected objective, we establish a convergence guarantee in terms of the first-order KKT residual of the original constrained problem. In particular, the proposed method computes an $\epsilon$-stationary solution in expectation within $O(\epsilon^{-4})$ stochastic gradient evaluations. Numerical experiments further show that the proposed method achieves competitive iteration complexity and improved wall-clock efficiency compared with representative recursive-momentum baselines.
Read the full abstract

Dynamical Systems

Global attractors and fast-slow reduction for finite-state actor-critic mean dynamics

When a learning algorithm reshapes the data distribution it trains on, the long-run behavior depends on the joint evolution of the policy, the value estimate, and the data distribution. We study finite-state actor-critic mean dynamics on the enlarged phase space $(\theta,w,\mu)$, where $\theta$ is the actor parameter, $w$ is an auxiliary critic state, and $\mu$ is a state-law variable (the distribution over states induced by the current policy). The state-law coordinate follows the exact controlled-Markov equation $\delta \dot\mu = Q_\theta^*\mu$. Under a softmax actor with box confinement (a smooth proxy for parameter clipping), a uniformly coercive linear critic equation, and a Lipschitz generator family $\theta \mapsto Q_\theta$, we prove that for each $\delta > 0$ the resulting autonomous semiflow possesses a compact global attractor. Under a uniform exponential-mixing assumption, we prove that the invariant-law map $\theta \mapsto \mu_\theta$ is Lipschitz and that the reduced invariant-law system on $(\theta,w)$ is well posed. Under an additional pathwise exponential-stability estimate for the non-autonomous fast state equation, we show that the exact flow tracks the reduced flow on every finite time interval up to the initial layer, and that the exact attractors converge upper semicontinuously to the lifted reduced attractor as $\delta \to 0$. We also give a concrete finite-state reference-state minorization condition implying the pathwise hypothesis. All results are formalized in Lean 4 without custom axioms.
Read the full abstract

Mathematical Physics

Continuation of Hamiltonian dynamics from the plane to constant-curvature surfaces

We investigate the deformation of symmetry on cotangent bundles from the Euclidean plane to two-dimensional constant-curvature surfaces and the continuation of local dynamics aspects in Hamiltonian systems. For a fixed curvature sign $\sigma\in\{+1,-1\}$, the curved problem is set up either on the sphere $(\sigma=+1)$ or on the hyperbolic plane $(\sigma=-1)$, both with radius $R=1/\varepsilon$, recovering flat space in the limit $\varepsilon\to 0$. The symmetry of these spaces is taken into account by using the Inönü–Wigner contraction of Lie algebras from $\mathfrak{so}(3)$ or $\mathfrak{so}(2,1)$ to $\mathfrak{se}(2)$. We use Riemannian exponential coordinates centred at the North pole together with the pull-back the associated momentum map and the symplectic form. Within this geometric setting we use a local slice construction and prove the persistence from flat to curved spaces of non-degenerate relative equilibria and relative periodic orbits of general cotangent bundle Hamiltonian systems. We apply the resulting framework to the Newtonian $n$-body problem.
Read the full abstract

Analysis of PDEs

Quantitative Kröger inequalities for Neumann eigenvalues of convex domains

Refining the sharp upper bounds $\mu_{k,d}^* $ obtained by Kröger (1999) for the $k$-th Neumann eigenvalue of a convex domain $\Omega \subset \mathbb{R}^d$, we prove the following inequalities: for any $k\in \mathbb{N}$ there exists a constant $C(k,d) >0$ such that $$D_{\Omega}^2 \mu_k(\Omega) \leq \mu_{k,d}^* - C(k,d) a_2(\Omega)^2/D_{\Omega}^2$$ where $D_{\Omega}$ is the diameter of $\Omega$ and $a_2(\Omega)$ is the second largest semiaxis of the John ellipsoid of $\Omega$. In the planar case, for $k=1$ we also give an explicit value of the constant $C(1,2)$.
Read the full abstract

Combinatorics

A quadratic form generalization of rational dinv

We introduce a quadratic form $Q$ on the space of functions on the gap poset $G$ of the numerical semigroup $\langle a,b\rangle$. We prove combinatorially that when evaluated on the indicator function of an upward closed subset $D$, this quadratic form precisely recovers the Gorsky–Mazin $\mathtt{dinv}$ statistic of $D$, viewed as a Young subdiagram of $G$. Furthermore, we prove Theorem~1.2 that when evaluated on a pair of subdiagrams of $G$, the symmetric bilinear form associated with $Q$ is equal to a novel cross-$\mathtt{dinv}$ statistic, which is nonnegative. Combining these, we prove the inequality \[ Q(\mathbf{n})\geq \dfrac{1}{|G|}\,\|\mathbf{n}\|_\infty^2\] if $\mathbf{n}$ is a real-valued decreasing function on $G$, showing an effective positive definiteness of $Q$ on the corresponding cone. Theorem~1.2, the main engine of the paper, was autoformalized in Lean/Mathlib by AxiomProver.
Read the full abstract

Numerical Analysis

Inverse scattering beyond Born approximation via rotation-equivariance-aware neural network and low-rank structure

This work proposes a hybrid method (ULR) which integrates a rotation-equivariance-aware neural network and a low-rank structure to solve the two dimensional inverse medium scattering problem. The neural network is to model the data corrector which maps the full data to the Born data, and the low-rank structure is to design an inverse Born solver that finds a regularized solution from the perturbed Born data. The proposed rotation-equivariance-aware neural network naturally incorporates the reciprocity relation and the rotation-equivariance in inverse scattering, while the low-rank structure effectively filters high-frequency noise in the output of the neural network and leads to a regularized method supported by theoretical stability in the Born region. For a comparative study, we replace the low-rank inverse Born solver by another rotation-equvariance-aware neural network to propose a two-step neural network (UU). Furthermore, we extend the proposed methods (ULR and UU) to tackle the more challenging case with only limited aperture data. A variety of numerical experiments are conducted to compare the proposed ULR, UU, and a black-box neural network.
Read the full abstract

Combinatorics

On derivatives and higher-order derivatives of chromatic polynomials

Let \( G \) be a graph of order \( n \) with maximum degree $\Delta$, and let $P(G,x)$ denote its chromatic polynomial. We investigate several properties of $P(G,x)$ related to its derivatives and higher-order derivatives. First, we study the monotonicity of $P(G,x)/x^n$. Dong proved that $(x-1)^nP(G,x)\geq x^nP(G,x-1)$ for all real $x\geq n$. In particular, taking $x=n$ establishes the Bartels-Welsh “shameful conjecture" that $P(G,n)/P(G,n-1)>e$. Fadnavis later showed that the same inequality holds for all real $x\geq 36\Delta^{3/2}$. We improve this bound by proving that it also holds for all real $x\geq 10\Delta^{3/2}$. We then consider a conjecture of Dong, Ge, Gong, Ning, Ouyang, and Tay asserting that \( \frac{d^k}{dx^k} \bigl( \ln[(-1)^n P(G, x)] \bigr) < 0 \) for all \( k \geq 2 \) and \( x \in (-\infty, 0) \). We establish this conjecture for all \( k \geq 2 \) and \( x\leq -3.01\Delta k \).
Read the full abstract

Statistics Theory · Cross-listed from stat.ML

Identifiability of Potentially Degenerate Gaussian Mixture Models With Piecewise Affine Mixing

Causal representation learning (CRL) aims to identify the underlying latent variables from high-dimensional observations, even when variables are dependent with each other. We study this problem for latent variables that follow a potentially degenerate Gaussian mixture distribution and that are only observed through the transformation via a piecewise affine mixing function. We provide a series of progressively stronger identifiability results for this challenging setting in which the probability density functions are ill-defined because of the potential degeneracy. For identifiability up to permutation and scaling, we leverage a sparsity regularization on the learned representation. Based on our theoretical results, we propose a two-stage method to estimate the latent variables by enforcing sparsity and Gaussianity in the learned representations. Experiments on synthetic and image data highlight our method's effectiveness in recovering the ground-truth latent variables.
Read the full abstract

Functional Analysis

Nonlocal Fourier Laws for Heat Propagation via Fractional powers of Vector Operators

The present work is devoted to the study of fractional powers of vector operators, with particular emphasis on the gradient operator with non-constant coefficients. Within the setting of Clifford algebra $\mathbb{R}_n$, this operator turns out to have bisectorial properties. By applying the spectral theory on the $S$-spectrum, we address a fundamental mathematical challenge: unlike sectorial operators, bisectorial operators involve fractional powers that are not analytic on the negative real line. To circumvent this, we introduce a novel definition of the fractional power function in this setting. Building upon previous works on bisectorial vector operators and weak solutions, we extend the definition of fractional powers to abstract vector operators. The core contribution of this work is the application of the functional calculus for vector operators to the gradient operator, showing that these fractional powers provide a rigorous mathematical foundation for nonlocal Fourier laws in heat propagation.
Read the full abstract