Probability
Signature Kernel and Schwinger-Dyson Kernel Equations as Two-Parameter Rough Differential Equations
Thomas Cass, Dan Crisan, Andrea Iannucci et al. (+1 more)
We develop a rough-path framework for two-parameter rough differential equations on rectangular and simplicial domains, motivated by the signature kernel and Schwinger–Dyson kernel equations. The theory is formulated in spaces of jointly controlled rough paths and is based on a robust two-parameter rough integration framework. In particular, we introduce a notion of rough integration over two-dimensional simplices at low regularity extending previous results in the literature. Within this setting, we show that the signature kernel equation arises naturally as a two-parameter rough differential equation and establish well-posedness and stability. We also extend the Schwinger–Dyson kernel equation, previously formulated for bounded-variation paths, to rough driving signals, proving existence and uniqueness in appropriate controlled rough path spaces. In the smooth rough path regime, we relate the resulting equations to PDE and integro-differential formulations. Finally, we derive and analyse a numerical scheme for the rough Schwinger–Dyson equation, including runtime and
memory complexity estimates, and illustrate its performance with numerical experiments.
Read the full abstract ▾
All authors Thomas Cass, Dan Crisan, Andrea Iannucci, William F. Turner
Functional Analysis · Cross-listed from quant-ph
Recoverable states on von-Neumann algebras
Saptak Bhattacharya
Let $(\mathcal{M},\tau)$ and $(\mathcal{N},\tau^{\prime})$ be tracial von-Neumann
algebras and let $\phi:\mathcal{M}\to\mathcal{N}$ be a strictly completely positive, trace preserving map. Given a positive, invertible $B\in\mathcal{M}$ with $\tau(B)=1$, a state on $\mathcal{M}$ given by a positive $A\in L^1(\mathcal{M}, \tau)$ is said to be recoverable if $\mathcal{R}(\phi(A))=A$ where $\mathcal{R}$ is the Petz recovery map corresponding to $B$ and $\phi$. In this paper, we study recoverable states and show how an arbitrary state can be made close to a recoverable state via iterates of $\mathcal{R}\circ\phi$. We show that there exists a completely positive, trace preserving map $\psi:\mathcal{M}\to\mathcal{M}$ such that $\psi(A)$ is recoverable for all $A$ and $(\mathcal{R}\circ\phi)^n\to\psi$ in norm as operators on $L^p(\mathcal{M},\tau)$ for all $1\,\textless p\,\textless\infty$, and discuss potential applications to
quantum information theory. We also show that this convergence holds strongly in $L^1$. Finally, we prove an interesting decomposition
theorem for normal states on $\mathcal{M}$.
Read the full abstract ▾
Information Theory
Fundamental Trade-Offs in Multi-Bit Watermarking of Stochastic Processes
Haiyun He, Yepeng Liu, Zhuoer Shen et al. (+3 more)
We study multi-bit watermarking for data generated by
stochastic processes, where a hidden message is embedded during sampling and must be decodable by an authorized detector that possesses side information unavailable to unauthorized observers. In high-stakes deployments, a practical watermark must simultaneously control false alarms, preserve generation quality without distorting the output distribution, and support reliable multi-bit decoding. Satisfying all three goals at once inevitably creates fundamental trade-offs. We formulate watermark
embedding as a distributional information-
embedding problem and watermark detection as a multiple-hypothesis testing problem under distortion and rate constraints, leading to four fundamental metrics: false-alarm probability, detection error probability, distortion, and information rate. Within this information-theoretic framework, we derive matched converse and achievability bounds that characterize the optimal trade-offs and provide scheme-agnostic
benchmarks for any watermarking method. For stationary ergodic
stochastic processes, we further obtain matched asymptotic limits and connect them to the finite-sample regime. Finally, we present a reference watermarking construction satisfying our assumptions and empirically illustrating the predicted trade-offs.
Read the full abstract ▾
All authors Haiyun He, Yepeng Liu, Zhuoer Shen, Ziqiao Wang, Yongyi Mao, Yuheng Bu
Mathematical Physics · Cross-listed from hep-th
Bogomol'nyi Equations in Mixed Product Chern-Simons Theories Governing Charged Vortices and Antivortices
Aonan Xu
We extend product Chern-Simons theory to develop several mixed $U(1)\times U(1)$ models where one gauge field is governed by a Chern-Simons term and the other by a Maxwell or Born-Infeld term. We show that, by choosing suitable potentials, the energy functional admits a topological lower bound saturated by first-order self-dual equations. The resulting dyonic systems can be divided into
vortex-
vortex and
vortex-antivortex configurations, and the coexistence of
vortices and antivortices in the latter extends the
vortex-only result known in product Chern-Simons model. On a doubly periodic domain, we establish Bradlow-type bounds with distinct physical implications: for
vortex-only systems, the
vortex numbers stay below this bound and cannot be arbitrarily large; for
vortex-antivortex systems, the bound is imposed on the difference between the
vortex and antivortex numbers, while the individual numbers are arbitrary. This distinction results in a bounded energy spectrum for the former and an unbounded energy spectrum for the latter.
Read the full abstract ▾
Classical Analysis and ODEs
Isoperimetric Inequality for degenerate elliptic operators of Grushin type
Dangyang He
Let $n,m\ge 1$, $\alpha\in(0,1)$, and $\beta\ge 0$. For the Grushin-type operator \[ L=-\nabla_x\!\cdot\!\bigl(|x|^{2\alpha}\nabla_x\bigr)+|x|^{2\beta}\Delta_y \qquad \text{on } \mathbb R^n\times \mathbb R^m, \] we prove the isoperimetric inequality on the associated Grushin space. Equivalently, if \[ Q=\frac{n+m(\beta+1-\alpha)}{1-\alpha}, \] then \[ |\Omega|^{\frac{Q-1}{Q}}\le C\,P(\Omega) \] for every smooth bounded domain $\Omega\subset \mathbb R^{n+m}$.
Read the full abstract ▾
Dynamical Systems
Fixed-point approximation for self-consistent transfer operators with Newton's method
Wael Bahsoun, Gary Froyland and Maxence Phalempin
Self consistent transfer operators arise naturally in the study of mean-field coupled dynamical systems and are closely related to kinetic PDEs such as the Vlasov equation. Despite substantial progress on existence and uniqueness of fixed points for self-consistent transfer operators, the development of fast, reliable, and provably accurate numerical methods remains largely unresolved. In this work, we construct a nonlinear
Fourier-Fejér discretisation and establish convergence of the resulting finite-dimensional fixed point to that of the original self-consistent transfer operator. Further, using the nonlinear
Fourier-Fejér discretisation, we prove exponential convergence of a sequential iteration scheme and develop a Newton framework with quadratic convergence. We present numerical examples demonstrating the efficiency and flexibility of the above methods.
Read the full abstract ▾
Differential Geometry
On weighted extremal Kähler metrics
Akito Futaki
The notion of weighted extremal Kähler metrics extends the classical notion of Calabi's extremal Kähler metrics, but includes many well-studied objects in Kähler geometry such as Kähler-
Ricci solitons and Sasaki-Einstein metrics. In this paper, after explaining how this notion grew out, we will try to survey recent works concerning the YTD
conjecture on weighted extremal Kähler metrics.
Read the full abstract ▾
Differential Geometry
On coupled Kähler-Einstein metrics and weighted solitons on Fano manifolds
Akito Futaki
We consider coupled Kähler-Einstein metrics and weighted
solitons on Fano
manifolds. These are natural generalizations of Kähler-Einstein metrics. As in the case of Kähler-Einstein metrics, the existence is known to be equivalent to algebraic conditions which generalize the K-polystability. In this survey, we outline recent developments for these two cases.
Read the full abstract ▾
Probability
Edge-averaging dynamics on finite graphs: moment dependence
Junchi Zuo
We study the edge-averaging process on a finite, connected graph $G = (V, E)$. Initially, the vertices in $V$ are endowed with i.i.d.\ real-valued opinions $(f_0(v))_{v \in V}$. Edges are activated according to i.i.d.\ Poisson clocks of rate $1$; when an edge is activated, the opinions at its endpoints are replaced by their average. Let $f_t(v)$ denote the opinion at $v$ at time $t$.Define the $\epsilon$-convergence time $\tau_\epsilon$ as the first time when the maximum and the minimum of $f_t$ differ by at most $\epsilon$. It is known that if the initial opinions $(f_0(v))_{v \in V}$ are bounded in $L^\infty$, then $\mathbb{E}(\tau_\epsilon)$ is at most $C_\epsilon \log^2 n$ for $\epsilon \in (0, 1]$. We assume instead that the $L^p$ norm of $f_0(v)$ is at most $1$ for every $v \in V$. For fixed $\epsilon \in (0, 1]$, and show that $\mathbb{E}(\tau_\epsilon) = \widetilde{O}(n^{\beta_p})$ up to logarithmic terms, where $\beta_p := \max(3 - p, 2/p)$. Moreover, this power law is tight on cycle graphs.
Read the full abstract ▾
Probability · Cross-listed from stat.ML
Measuring and Decomposing Mode Separation via the Canonical Diffusion
Shaul Tolkovsky, Ori Meidler and Or Zuk
Mode separation, namely how sharply a distribution fragments into barrier-separated clusters, is a fundamental
geometric property of densities, difficult to quantify in high dimensions. Modes are often the meaningful alternatives in the data, such as the conformations of a molecule or the distinct semantic types among a generative model's outputs. Mode separation is distinct from dispersion, which is what standard label-free tools capture: differential
entropy rises with spread regardless of fragmentation, and PCA orders directions by variance regardless of barriers. Mutual information does capture it, but only given the latent component labels, which are usually unavailable. We measure mode separation through a single
stochastic process intrinsic to the density, namely the unique reversible
diffusion with $f$ as its stationary distribution and a constant scalar
diffusion coefficient. From its autocovariance
matrix we extract two readouts. The first, SSA (Sum of Squared Autocorrelations), is a scalar barrier-sensitive measure. The second, DA (Dominant Autocorrelation directions), gives linear projections ordered by metastability rather than variance. Under an isotropic-Gaussian null, we derive the limiting spectrum of the empirical autocovariance in closed form, generalizing the Marchenko–Pastur law. Its upper edge gives a noise floor that separates metastable directions from sampling noise. Both readouts use only samples and a score function, scaling to high dimensions through pretrained score-based generative models via Tweedie's identity. We apply our framework to three settings: (i) synthetic Gaussian mixtures, where SSA tracks mutual information; (ii) SDXL text-to-image generations, where SSA and DA capture structure that
entropy and PCA miss; and (iii) molecular dynamics of alanine dipeptide, where DA recovers the known slow backbone dihedrals from static samples alone.
Read the full abstract ▾
Optimization and Control · Cross-listed from cs.RO
Omni-scale Learning-based Sequential Decision Framework for Order Fulfillment of Tote-handling Robotic Systems
Jiaxin Liu, Peng Yang, Yuping Li et al. (+1 more)
Driven by the rapid expansion of e-commerce and small-batch production, the size of the intralogistics load unit of finished goods, semi-finished goods and raw materials is steadily shrinking. Totes are gradually replacing pallets as the primary handling and storage container. This shift has propelled tote-handling robotic systems to the forefront of automation order fulfillment centers. The order-fulfillment decisions of tote-handling robotic systems share a common order-tote-
robot sequential decision-making nature. Existing studies primarily focus on decision mechanisms tailored to particular systems, making it difficult to generalize or transfer them to other contexts. We propose an Omni-scale Learning-based Sequential Decision Framework for Order Fulfillment of Tote-handling Robotic Systems (OLSF-TRS), a generalized and scalable sequential decision framework that combines structured combinatorial optimization with multi-
agent reinforcement learning to coordinate order,tote, and
robot decisions. On small-scale tote-handling robotic systems, OLSF-TRS achieves near-optimal performance with average optimality gaps below 3.5% across two distinct system configurations. In large-scale scenarios, OLSF-TRS consistently outperforms heuristic baselines across two different system types, reducing total tote movements by 8-12% and over 30% compared to SOTA rule-based approaches, while maintaining real-time responsiveness. These improvements translate into tangible operational benefits, including cost reduction, lower energy consumption, and enhanced throughput stability. The proposed framework delivers an efficient and unified order fulfillment decision-making framework for widely deployed tote-handling robotic systems,supporting high-quality order fulfillment in both e-commerce and industrial logistics sectors.
Read the full abstract ▾
All authors Jiaxin Liu, Peng Yang, Yuping Li, Xinyue Xie
Optimization and Control
On Composite Adaptive Continuous Finite-Time Control of a class of Euler-Lagrange systems
Emmanuel Cruz-Zavala, Jaime A. Moreno and Antonio Loría
In this paper, we propose several set-point control schemes for achieving finite-time regulation in a class of Euler–Lagrange systems with $n$ degrees of freedom and uncertain potential energy. The proposed controllers are based on composite adaptive control approaches. Each control scheme consists of two main components: a Proportional–Derivative (PD)-based nonlinear feedback term and a finite-time parameter estimation law. The estimation laws rely on the Dynamic Regressor Extension and Mixing (DREM) technique, which can be designed using either the Kreisselmeier or the least-squares dynamic regressor extensions. These results extend recent advances in finite-time adaptive control for Euler-Lagrange systems. To the best of the authors' knowledge, the composite adaptive control formulation proposed here, which does not employ well-known Slotine and Li adaptive control structure, has not been studied in detail yet. The properties of the proposed controllers are rigorously analyzed using Lyapunov methods. The performance of the controllers is thoroughly evaluated through extensive simulation studies.
Read the full abstract ▾
Complex Variables
Uniqueness of entire functions sharing two values with their partial derivative operators
Sujoy Majumder, Debabrata Pramanik and Shantanu Panja
In this paper, we employ the theory of normal families in several complex variables to obtain some uniqueness
theorems for entire functions. These results extend the related works of Li and Yi [11], and Lu et al. [18] to the setting of several complex variables. Moreover, some examples are provided to demonstrate the sharpness of our results.
Read the full abstract ▾
Dynamical Systems · Cross-listed from cs.LG
The Global Empirical NTK: Self-Referential Bias and Dimensionality of Gradient Descent Learning
James Hazelden, Laura Driscoll, Eli Shlizerman et al. (+1 more)
In training a neural network with gradient descent (GD), each iteration induces a linear operator that governs the first-order updates to a model's internal state variables. We call this operator the Global Empirical Neural Tangent Kernel (GeNTK). In finite-width networks, GeNTK is typically intractable to form, leading prior work to focus on restrictive settings, such as tracking only outputs or infinite-width limits. Here, we bridge this gap by studying the structure of GeNTK across many models. Formulating the model state as the solution to a global implicit constraint, we derive GeNTK as a product of two operators: K, accounting for immediate parameter-to-state interactions, and P, describing internal state-to-state dependencies. For a broad class of "weight-based models", such as RNNs and
transformers, we prove a universal Kronecker-core
theorem showing that K admits an exact, forward-pass-computable form given by the Gram
matrix of weight-site variables. This core gives a measurable state-dependent restriction on Euclidean GD, termed "self-referential bias": current activity filters which adjoint directions can drive a first-order update. For recurrent models (GRUs and RNNs), we examine the spectrum of GeNTK and formalize notions of low-rank, biased learning in terms of reduced spatial or temporal ranks of the full operator. Practically, this explains why some RNN initializations yield better learning, while some plateau, by biasing the GeNTK and thus restricting learning to a small domain of learnable tasks. To demonstrate computational viability, we validate the main
theorems in a self-
attention transformer and a pretrained Pythia
transformer language model with 410 million parameters. Overall, our results expose tractable local rank and mode-selective biases in the finite-width GD geometry and open the door to principled study of the GeNTK, a bridge between theory and practice.
Read the full abstract ▾
All authors James Hazelden, Laura Driscoll, Eli Shlizerman, Eric Shea-Brown
Information Theory · Cross-listed from cs.LG
Generative Actor-Critic with Soft Bridge Policies
Ke He, Le He, Shunpu Tang et al. (+2 more)
Expressive generative policies such as
diffusion and flow models are appealing for maximum-
entropy (MaxEnt) online reinforcement learning because of their ability to model
multimodal and highly non-Gaussian action distributions. However, training effective soft generative policies faces two obstacles that often arise together. First, marginal action densities are often unavailable, so existing methods typically rely on
entropy bounds, heuristic proxies or approximations. Second, iterative shared-parameter samplers raise inference cost and require backpropagation through time over repeated network evaluations, increasing
memory cost and potentially destabilizing policy optimization. These obstacles motivate us to seek a generative actor-critic method that exposes a tractable soft objective while generating each action in just one network evaluation. To this end, we propose soft generative actor-critic (SoftGAC), whose actor is a single
stochastic feedforward network with internal Gaussian bridge layers. This explicit bridge structure allows us to lift the soft objective to path space, yielding an analytical path-KL regularizer against a high-
entropy reference process for both actor improvement and the soft Bellman bootstrap, without estimating the endpoint
entropy. Crucially, the actor generates each action in only one network evaluation under a total parameter budget comparable to strong baselines, substantially reducing inference cost relative to the existing iterative
diffusion and flow policies. It also avoids long unrolled backpropagation through time. Across challenging continuous-control
benchmarks, our method achieves competitive or superior returns compared with strong
diffusion and flow policies, demonstrating a clear advantage in the compute-return tradeoff.
Read the full abstract ▾
All authors Ke He, Le He, Shunpu Tang, Yafei Wang, Lisheng Fan
Functional Analysis
A constructive approach to the truncated moment problem on cubic curves in Weierstrass form
Abhishek Bhardwaj and Aljaž Zalar
In this paper, we develop a constructive solution for the pure truncated moment problem on cubic curves in Weierstrass form, establishing the existence of a representing measure whose number of atoms equals the rank of the associated moment
matrix. By a recent result of Baldi, Blekherman, and Sinn, for projectively smooth curves whose projective closure has exactly one real point at infinity, the existence of such a rank-attaining atomic measure is equivalent to the existence of a representing measure; consequently, the TMP is constructively solved for this class of curves. We also present a numerical degree–$6$ example in which every minimal representing measure supported on the cubic curve requires $\operatorname{rank} M(3)+1$ atoms, where $M(3)$ denotes the moment
matrix. Finally, we provide a constructive solution for the symmetric case, i.e., when all moments of odd degree in $y$ vanish.
Read the full abstract ▾
Information Theory
Sensing-Aided Secure Multicast in Rotatable Antenna-Enabled ISAC Systems
Zequan Wang, Liang Yin, Chongjun Ouyang et al. (+4 more)
Acquiring the channel state information (CSI) of passive eavesdroppers remains a fundamental challenge in physical layer security. The sensing capability of integrated sensing and communication (
ISAC) systems enables estimation of a potential eavesdropper's angle of departure (AoD) before secure transmission. Accordingly, a sensing-aided secure multicast scheme is proposed using a rotatable
antenna (RA) architecture that combines array-level and element-level rotations with analog beamforming. The scheme comprises eavesdropper sensing and secure communication stages. In the sensing stage, the maximum likelihood
estimator (MLE) and corresponding Cramer–Rao bound (CRB) are derived for eavesdropper AoD estimation. The two rotation levels are then optimized through cyclic coordinate search to minimize the worst-case CRB. The resulting AoD estimate and CRB determine the center and width of the angular uncertainty region, respectively. In the communication stage, the constant-modulus analog beamformer and RA configuration are jointly optimized to maximize the worst-case secrecy rate over this region. After angular discretization and smooth approximation, the resulting problem is solved using a product-space joint optimization framework. Numerical simulation results validate the convergence and effectiveness of the proposed
algorithms. It is demonstrated that i) the proposed RA-enabled sensing design effectively improves the eavesdropper AoD estimation accuracy; ii) a high and nearly constant secrecy rate is maintained over the uncertainty region; and iii) the joint optimization of the two rotation levels yields lower CRBs and higher secrecy rates than schemes employing either a fixed-position array or a single rotation level.
Read the full abstract ▾
All authors Zequan Wang, Liang Yin, Chongjun Ouyang, Hao Xu, Yunan Sun, Yitong Liu, Hongwen Yang
Analysis of PDEs
Well-posedness and regularity for seminlinear time-dependent second and fourth order in space equations
Gopikrishnan Chirappurathu Remesan
This article discusses a unified convergence analysis of the semilinear time-dependent equation $\partial_t u + (-1)^\mathrm{m}\Delta^{\mathrm{m}}u + u^3 - u = f$ with $\mathrm{m} \in \{1,2\}$ and homogeneous Dirichlet boundary conditions. The analysis relies on Faedo-Galerkin approximation and convergence via compactness estimates. The existence and uniqueness of the weak solution is proved when the initial data is smooth. A refined and novel analysis extends the existence result to problems with rough initial data also.
Read the full abstract ▾
Algebraic Topology
Homotopy exponents of polyhedral products
Briony Eldridge
We study Moore's
conjecture and homotopy exponents for polyhedral products. For $(\underline{CA},\underline{A})^K$ where each $A_i$ is finite and has torsion-free homology, we prove that if $(\underline{CA},\underline{A})^K$ is rationally
hyperbolic, then it has no homotopy exponent at any odd
prime. Under the additional hypothesis $\Sigma A_i$ is homotopy equivalent to a finite-type wedge of simply-connected spheres, we show Moore's
conjecture holds for $(\underline{CA},\underline{A})^K$. We also give criteria such that, for a large family of polyhedral join products, the associated polyhedral products are rationally
hyperbolic, mod-$p^r$
hyperbolic for all but finitely many
primes, and have no homotopy exponent at all but finitely many
primes.
Read the full abstract ▾
Probability
Normal approximation of the numbers of isolated edges and isolated 2-stars in uniform simple graphs with given vertex degrees
Ryo Imai
We consider the configuration model and the uniform simple graph with given degree sequence $\boldsymbol{d}=\left(d_i\right)_{i=1}^n$. We derive quantitative bounds for the errors in (i) joint normal-Poisson approximation to the numbers of isolated edges, isolated 2-
stars, self-loops and double edges in the configuration model, and (ii) normal approximation to the numbers of isolated edges and isolated 2-
stars conditioned on that the configuration model is simple. The latter provides the first finite sample normal approximation results for the uniform simple graph with given vertex degrees. To achieve this, we develop a new Stein's method for joint normal-Poisson approximation and a new coupling approach to sums of indicators, which may be of independent interest.
Read the full abstract ▾
Statistics Theory
Minimax Optimal Estimation of Transport-Growth Pairs in Unbalanced Optimal Transport
Donlapark Ponnoprat, Noboru Isobe and Masaaki Imaizumi
Unbalanced optimal transport extends classical optimal transport to measures with different total masses, but statistical guarantees for estimation of the transport map and the growth map remain limited. We study unbalanced optimal transport with quadratic cost and Kullback-Leibler marginal penalties and argue that the natural target of estimation is a transport-growth pair. Consequently, we develop two
estimators for the transport-growth pair under two setups: a nearest-neighbor
estimator for transporting between general densities, and a kernel-based
estimator for transporting between smooth densities. We prove that the kernel-based
estimator achieves the minimax optimal rate up to a logarithmic factor by deriving a matching lower bound on the minimax risk. Our main technical contribution is a reduction of estimation to a stability bound on the unbalanced optimal transport value. This bound controls the transport map and growth map risks through the duality gap of the true dual potentials for the estimated measures. These results provide a statistical foundation for estimation of the transport-growth pair in unbalanced optimal transport.
Read the full abstract ▾
Analysis of PDEs
The Ekeland–Nirenberg Variational Problem:A Sharp Positivity Threshold and Extensions
Qi Guo, Xueping Huang and Yi Huang
We study the Ekeland–Nirenberg variational problem in the two-dimensional diagonal family \[ J_{a,c,d}(u)=\int_{\Rp^2}\bigl(u_{xy}^2+a u_x^2+c u_y^2+d u^2\bigr)\dd x\dd y, \qquad a,c,d>0, \] under the constraint $u(0,0)=1$. If $u_{a,c,d}$ is the unique minimizer and $K_{a,c,d}$ is its cosine kernel, we prove the sharp classification \[ K_{a,c,d}>0 \hbox{ on } \Rp^2\quad\Longleftrightarrow\quad u_{a,c,d}>0 \hbox{ on } \Rp^2\quad\Longleftrightarrow\quad d\le ac . \] Thus every supercritical triple $d>ac$ produces sign change. We also prove local sign-change stability under small two-dimensional non-diagonal perturbations and a sharp product-type $n$-dimensional diagonal threshold. The domain and
evolution results are stated in precise auxiliary settings: a free-boundary capacity formulation for domains and a selected decaying branch of the second-order
evolution equation.
Read the full abstract ▾
Information Theory · Cross-listed from cs.CR
AI-Accelerated Brute Force Cryptanalysis
Gideon Samid
Modern cryptography is hinged on "not learning from mistakes": trying numerous wrong keys, should not help one identify the right key. Indeed, it worked – until recently when the surprising power of AI to see pattern in apparent randomness has turned the 'wrong plaintexts' generated by the 'wrong key' into productive inferential input. Crunching through these random-looking plaintext candidates AI can de-flatten the probability curve over the remaining key space. The more spiked this curve, the faster the ciphertext is defeated. This new attack vector demands a thorough review of our cryptographic security posture. NIST PQC is not immunized against AI-Accelerated Brute Force attack. Defense is rooted in non-trivial ciphertexts, in unilateral randomness, and in variable key size. This points to a new security class: Pattern Devoid Cryptography which is to be added into the toolbox used by the cyber security community.
Read the full abstract ▾
Optimization and Control
On the Stationary Duality of Structural Composite Cardinality Optimization
Penghe Zhang, Naihua Xiu and Houduo Qi
Simple cardinality refers to counting nonzero elements of an independent variable satisfying certain properties. Composite cardinality is a simple counting process composited with an affine mapping, and is therefore more complicated than the simple cardinality. We study the composite cardinality optimization problem (CCOP) with structures covering a wide range of applications. Through the use of the stationary duality, we reduce the composite counting to simple counting, and thereby obtain a dual formulation of CCOP. For both primal and dual problems, we investigate the sufficient conditions for the existence of global solutions. Those conditions are validated on representative examples from existing literature. We then show that local solutions of the primal and dual problems are equivalent to their stationary points. This result further helps us establish a one-to-one correspondences between primal and dual local solutions. We also demonstrate that the correspondence holds for a pair of global solutions to the primal and dual problems, provided that the dual weighted parameters are appropriately selected. The reported theoretical results lay foundation for developing numerical
algorithms for CCOP in future.
Read the full abstract ▾
Numerical Analysis · Cross-listed from stat.ML
Core-Halo Decomposition: Decentralizing Large-Scale Fixed-Point Problems
Haixiang, Yang Xu, Jiefu Zhang et al. (+4 more)
We study solving large-scale fixed-point equation \(x^\star=\bar F(x^\star)\) with decomposition. Standard strict decomposition assigns each
agent a disjoint block and evaluates updates using only owned coordinates. For most operators, however, a block update may depend on variables outside the block. Truncating these dependencies by strict decomposition changes the mean operator and creates structural bias that cannot be removed by more samples, smaller stepsizes, or additional consensus. We therefore propose Core-
Halo decomposition, which separates write ownership from read-only evaluation context: each
agent updates its own core and reads from an overlapping
halo. By aligning the Core-
Halo decomposition with the block-dependence structure of $\bar F$, the original fixed-point problem can be implemented faithfully in a decentralized multi-
agent system. We further characterize the fundamental obstruction faced by strict decomposition through a Bellman closure condition and a blockwise bias lower bound, showing that local-only updates can alter the original fixed-point operator. Finally, we conduct extensive experiments across a range of application settings, and demonstrate that Core-
Halo achieves near-centralized performance while retaining the parallelism benefits of decentralization.
Read the full abstract ▾
All authors Haixiang, Yang Xu, Jiefu Zhang, Xudong Wu, Zihan Zhou, Jun He, Jiayu Chen
Statistics Theory
Bridging Theory and Practice: Statistical Inference for Latent Space Models of Networks
Yuang Tian, Jiajin Sun and Yinqiu He
Latent space models have been widely adopted in modeling network data. Developing statistical inference for estimated model parameters enables quantifying associated uncertainty and is pivotal for downstream tasks. Despite recent progress on statistical inference of maximum likelihood estimation, crucial gaps remain between asymptotic theoretical guarantees and practical use. Specifically, how are the oracle maximum likelihood
estimators related to the solutions produced by
algorithms in practice? Can rigorous guarantees be established for existing
algorithms without unnecessary restrictions? To address these fundamental questions, we develop a unified analytical framework that bridges theory and practice of statistical inference for latent space models. First, for the maximum likelihood estimation, we relax the spectral-multiplicity constraint in the existing asymptotic theory to broaden the applicability. Second, we overcome the dependence on unknown true parameters in prior algorithmic analyses by developing novel adaptive criteria and theoretical tools. For the widely used
algorithm based on the projected gradient descent and the singular value thresholding, we explicitly connect their outputs to the maximum likelihood
estimator without relying on unknown information. Our results provide a solid foundation for practically useful and statistically principled statistical inference in network analysis.
Read the full abstract ▾
Combinatorics
Moonflowers and efficient code sparsification
Shachar Lovett, Raghu Meka and Yimeng Wang
We introduce moonflowers, a weaker analogue of sunflowers. A family of sets $S_1,\ldots,S_k$ is a $k$-moonflower if each set $S_i$ contains at least one element that is absent from all the others. We study the extremal problem of determining the largest possible size of a family of sets of size at most $w$ that avoids a $k$-moonflower, and obtain near-optimal bounds. As an application, we revisit the code sparsification problem studied by Brakensiek and Guruswami (STOC 2025) and improve the bounds to near optimal. Concretely, we improve the dependence on the block length from poly-logarithmic to logarithmic, and show that such a dependence is necessary.
Read the full abstract ▾
Statistics Theory
Posterior Concentration of Bayesian Physics-Informed Neural Networks for Elliptic PDEs
Yuxuan Zhao and Yulong Lu
We study the posterior contraction rate of
Bayesian Physics-Informed Neural Networks (PINNs) for solving a general class of
elliptic partial differential equations (PDEs). We focus on learning of the
elliptic equation with a non-homogeneous Dirichlet boundary condition from independent and noisy measurements collected both inside the domain and on the boundary. Assuming that the PDE admits a strong solution in a Hölder space and using with a suitably constructed prior on the neural network weights, we prove that the posterior distribution concentrates around the exact solution at a near-minimax rate. Furthermore, the chosen prior is rate-adaptive: the posterior contracts at an (almost) optimal rate without prior knowledge of the smoothness level of the exact solution. Our results provide statistical guarantees for uncertainty quantification of PDEs via
Bayesian PINNs.
Read the full abstract ▾
Combinatorics
A search for Hadamard matrices of Williamson type
Hadi Kharaghani, Ali Mohammadian and Behruz Tayfeh-Rezaie
In this article, we consider a special class of Williamson type
matrices which we call them near Williamson
matrices. They are in fact four $n\times n$ $(-1, 1)$-
matrices $A, B, C, D$ so that $A$ is circulant, $B,C,D$ are symmetric circulant, and they satisfy $AA^\top+BB^\top+CC^\top+DD^\top=4nI$. Using a computer search, we find all inequivalent near Williamson
matrices for all odd orders at most $35$. We also show that such
matrices exist for all odd orders up to $63$. As a consequence, we find the first known example of a quaternary Hadamard
matrix of order $118$.
Read the full abstract ▾
Combinatorics
On the multipliers of a Singer quadrangle
Wendi Di and Tao Feng
A finite generalized quadrangle $\cS$ is a Singer quadrangle if it has an
automorphism group that acts sharply transitively on its points. In this paper, we introduce the notion of multipliers for a Singer quadrangle and study their basic properties. As an application, we show that a point-primitive
automorphism group of a thick generalized quadrangle cannot have O'Nan-Scott type HS (holomorph simple), which answers an open problem in \cite{Bamberg 2019}.
Read the full abstract ▾