Peer Revenue Returns subject to peer review

Papers

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

Mathematics

math Watch Mathematics (sign in to use)

Tuesday 17 March

Optimization and Control

OT-DETECT: Optimal Transport-Driven Attack Detection in Cyber-Physical Systems

This letter presents an optimal-transport (OT)-driven, distributionally robust attack detection algorithm, OT-DETECT, for cyber-physical systems (CPS) modeled as partially observed linear stochastic systems. The underlying detection problem is formulated as a minmax optimization problem using 1-Wasserstein ambiguity sets constructed from observer residuals under both the nominal (attack-free) and attacked regimes, and show that the minmax detection problem can be reduced to a finite-dimensional linear program for computing the worst-case distribution (WCD). Off-support residuals are handled via a kernel-smoothed score function that drives a CUSUM procedure for sequential detection. We also establish a non-asymptotic tail bound on the false-positive error of the CUSUM statistic under the nominal (attack-free) condition, under mild assumptions. Numerical illustrations are provided to evaluate the robustness properties of OT-DETECT.
Read the full abstract

Differential Geometry

Collapsing Flat ${\rm{SU}}(2)$-Bundles to Spherical 3-Manifolds

We present a geometric mechanism for the emergence of spherical $3$-manifolds from the superspace of Riemannian metrics associated with flat ${\rm{SU}}(2)$-bundles over closed orientable hyperbolic surfaces. Our main result shows that any homogeneous spherical 3-manifold $(S,g_{S})$ can be realized as a boundary point in the Gromov-Hausdorff closure of a superspace $\mathcal{S}(P)$, where $P$ is a flat ${\rm{SU}}(2)$-bundle over a closed orientable hyperbolic surface $(\Sigma,h_{\Sigma})$. We show that the convergence of the sequence of metric spaces towards the spherical limit is controlled by the order of the fundamental group of $S$ and the metric invariant of the hyperbolic base provided by the ratio between its area and its systole. In this framework, the problem of obtaining the sharpest upper bound error reduces to the classical problem of maximizing the systole function over the moduli space of hyperbolic Riemann surfaces. As a byproduct, we observe that certain arithmetic surfaces provide the best possible error estimates within this family. To illustrate these results, we show that, according to our mechanism, the Bolza surface yields the optimal error bound for the convergence toward the Poincaré homology sphere.
Read the full abstract

Functional Analysis

A note on differentials of holomorphic functions

Recently, in arXiv:2304.07149, a bridge was made between the very active area of spaces of Lipschitz real functions on a metric space and holomorphic functions on an open subset of a Banach space. This was done by introducing and studying the space $\mathcal HL_0(B_X)$ of holomorphic Lipschitz functions defined on $B_X$, the open unit ball of the complex Banach space $X$ vanishing at 0. There it was proved that this space is isometrically isomorphic to a subspace of $\mathcal H^\infty(B_X, X^*)$, the space of bounded holomorphic mapping with values in the topological dual of $X$. In that paper it was shown that this subspace was a proper one, except in the one dimensional case. The goal of this note is to give an intrinsic characterization of the elements of that subspace. Moreover, in the case where $X$ additionally has a Schauder basis, it is shown that there is an explicit way to calculate whether and element of $\mathcal H^\infty(B_X, X^*)$ belongs or not to that subspace.
Read the full abstract

Optimization and Control

Preconditioned Proximal Gradient Methods with Conjugate Momentum: A Subspace Perspective

In this paper, we propose a descent method for composite optimization problems with linear operators. Specifically, we first design a structure-exploiting preconditioner tailored to the linear operator so that the resulting preconditioned proximal subproblem admits a closed-form solution through its dual formulation. However, such a structure-driven preconditioner may be poorly aligned with the local curvature of the smooth component, which can lead to slow practical convergence. To address this issue, we develop a subspace proximal Newton framework that incorporates curvature information within a low-dimensional subspace. At each iteration, the search direction is obtained by minimizing a proximal Newton model restricted to a two-dimensional subspace spanned by the current preconditioned proximal gradient direction and a momentum direction derived from the previous iterate. By orthogonalizing the subspace basis with respect to the local Hessian-induced metric, the solution of the original coupled two-dimensional nonsmooth subproblem can be well approximated by solving two decoupled one-dimensional subproblems, while keeping the per-iteration computational cost low. We establish global convergence of the proposed method and prove a $Q$-linear convergence rate under strong convexity. Comparative numerical experiments demonstrate the effectiveness of the proposed algorithm on ill-conditioned problems.
Read the full abstract

Numerical Analysis

Splitting horizontal and vertical polynomial order in a compatible finite element discretisation for numerical weather prediction

The accurate and efficient representation of atmospheric dynamics remains a central challenge in numerical weather prediction. A particular difficulty arises from the strong anisotropy of the atmosphere, in which horizontal and vertical motions occur on very different length scales, motivating numerical discretisations that can reflect this structure. In this study, we introduce a compatible finite element discretisation of the compressible Boussinesq and compressible Euler equations in which the horizontal and vertical polynomial orders are treated independently. The split-order discretisation is constructed using a tensor-product framework that preserves the discrete de Rham complex and associated mimetic properties. Its wave-propagation characteristics are examined through a discrete dispersion analysis that extends previous analyses to configurations with differing horizontal and vertical polynomial orders. The results show that increasing horizontal order improves the representation of gravity waves at low and intermediate wavenumbers, while increasing vertical order can degrade dispersion accuracy near the grid scale and introduce spectral gaps. A series of idealised numerical experiments, including gravity-wave propagation, advective transport, mountain-wave flow, and a global baroclinic-wave test, is used to assess the scheme's accuracy and convergence properties. These experiments demonstrate that increasing the polynomial order in the dominant direction of motion improves convergence, and that increasing the horizontal order yields the greatest gain in accuracy under typical atmospheric conditions. The results indicate that split-order compatible finite element discretisations provide a viable alternative for controlling accuracy and numerical behaviour in atmospheric dynamical cores.
Read the full abstract

Mathematical Physics · Cross-listed from quant-ph

Quantum classification and search algorithms using spinorial representations

We propose an algebraic formulation for two distinct quantum algorithms: a quantum classification algorithm and a quantum search algorithm with a non-uniform initial distribution, both based on Clifford algebras and spinorial representations. In the classification algorithm, we exploit properties of spinorial representations to construct orthogonal quantum states associated with different classes, allowing the identification of an item's class through the evaluation of expectation values of operators derived from the generators of the Clifford algebra. In the quantum search algorithm, we consider a database with prior information in which the oracle is implemented directly using generators of the Clifford algebra, simplifying its realization. The proposed approach provides a unified algebraic description for both algorithms, employing spinorial representations in the construction of quantum states and operators. Computational implementations are presented.
Read the full abstract

Combinatorics

Geometric constructions for Steinitz-type bounds in dimension two

We investigate inequalities for partial sums of complex numbers with bounded modulus and zero total sum, a topic referred to as "polygonal confinement". Starting from Steinitz's classical result, we provide detailed constructions yielding explicit bounds, including $\sqrt{5}$, $\sqrt{3}$, $2$, and $\sqrt{2}$, depending on geometric constraints or weighted settings. The proofs are fully detailed with step-by-step constructions of permutations, highlighting the combinatorial and geometric intuition. We conclude with conjectures on optimal universal constants and directions for future research.
Read the full abstract

Algebraic Geometry

1-motives and admissible variations of mixed Hodge structures

Let S be a connected scheme smooth and of finite type over the field of complex numbers. To every 1-motive over S, André associated the enriched Hodge realization given by a torsion-free, graded-polarizable and admissible variation of mixed Hodge structures of type (0,0), (-1,0), (0,-1), (-1,-1) over the associated complex analytic space. In this paper, we prove that every admissible variation of mixed Hodge structures of the above type arises, up to isogeny, from a 1-motive over S, thereby providing a positive answer to a question of André concerning the geometric origin of such variations. More precisely, we establish a Hodge-theoretic interpretation of sections of semi-abelian varieties by combining André's description of the abelian case with a new analysis of the toric part. As a consequence, we prove a relative analogue of Deligne's equivalence over the field of complex numbers. Namely, under suitable assumptions on S and on the lattices and the tori underlying 1-motives, the enriched Hodge realization functor induces an equivalence between the category of 1-motives over S and the category of torsion-free, graded-polarizable and admissible variations of mixed Hodge structures of type (0,0), (-1,0), (0,-1), (-1,-1). In general, the corresponding statement holds only up to isogeny. Finally, we introduce the global Mumford–Tate group of a 1-motive over S and show that its neutral connected component identifies with the Mumford-Tate group of the generic fiber.
Read the full abstract

Differential Geometry

Pohozaev identities and Liouville theorems for $L$-tension energies on manifolds

We study autonomous energies $B(\phi,e,\|\tau_L\|^2/2)$ for a general Lagrangian $L=L(y,r)$. The scaling derivative $\widehat L=2rL'$ determines the trace of the stress tensor and a vector-field identity on Riemannian domains. For homogeneous $L=a(y)(2r)^{p/2}/p$, the divergence term in this trace vanishes when $p=m$. A conformal vector field with a strictly signed conformal factor then gives $L$-harmonicity for positive mixed tension powers, under explicit integrability and flux conditions. This applies, in particular, to hyperbolic space. On complete gradient Ricci solitons with bounded Ricci curvature, a separate cutoff argument gives a global identity for general $L$. On the cigar and on Euclidean space, direct metric variations give second-order boundary criteria that also cover tension powers $1<q<2$. The applications include target weights and the fourth-power case. Examples distinguish the annular condition from global first-derivative integrability and show why scalar growth bounds or finite tension energy alone do not imply vanishing.
Read the full abstract

Differential Geometry

Nice bases for Lie algebras

The concept of a nice basis for a Lie algebra was introduced to study the Ricci curvature on nilpotent Lie groups equipped with a left-invariant metric. Despite the many applications in differential geometry, for example in the construction of Einstein manifolds, very little is known about the existence and number of nice bases on a given Lie algebra. This paper studies this question for three classes of Lie algebras, namely direct sums, almost abelian ones and nilpotent Lie algebras associated to a graph. As an application we compute the number of nice bases for Lie algebras up to dimension $3$, and show that for a general Lie algebra the existence depends on the field over which it is defined. Moreover, for every natural number $n$ we give an indecomposable Lie algebra such that there exists exactly $n$ nice bases up to equivalence.
Read the full abstract

Numerical Analysis

Perturbation Analysis of the QT-Drazin Inverse of Quaternion Tensors via the QT-Product

The motivation of this paper is to investigate the perturbation theory for the QT-Drazin inverse of quaternion tensors under the QT-product via the associated $z$-block circulant representation. A fundamental relationship between the QT-Drazin inverse of $\mathtt{bcirc}_z(\mathcal A)$ and the $z$-block circulant form of $\mathcal A^D$ is established. Moreover, the QT-index of a quaternion tensor is characterized by the indices of the diagonal blocks in the corresponding block-diagonalized matrix. As a consequence, a representation of the QT-Drazin inverse in terms of the QT-Moore–Penrose inverse is derived, which offers a practical approach for its direct computation in MATLAB. Furthermore, a decomposition theory for the QT-Drazin inverse is developed by combining the structure of $z$-block circulant matrices with the Jordan decomposition of quaternion matrices. Numerical examples are provided to demonstrate the theoretical results and computational feasibility.
Read the full abstract

All authors Yue Zhao, Daochang Zhang, Jingqian Li, Dijana Mosic

Optimization and Control · Cross-listed from cs.LG

SympFormer: Accelerated attention blocks via Inertial Dynamics on Density Manifolds

Transformers owe much of their empirical success in natural language processing to the self-attention blocks. Recent perspectives interpret attention blocks as interacting particle systems, whose mean-field limits correspond to gradient flows of interaction energy functionals on probability density spaces equipped with Wasserstein-$2$-type metrics. We extend this viewpoint by introducing accelerated attention blocks derived from inertial Nesterov-type dynamics on density spaces. In our proposed architecture, tokens carry both spatial (feature) and velocity variables. The time discretization and the approximation of accelerated density dynamics yield Hamiltonian momentum attention blocks, which constitute the proposed accelerated attention architectures. In particular, for linear self-attention, we show that the attention blocks approximate a Stein variational gradient flow, using a bilinear kernel, of a potential energy. In this setting, we prove that elliptically contoured probability distributions are preserved by the accelerated attention blocks. We present implementable particle-based algorithms and demonstrate that the proposed accelerated attention blocks converge faster than the classical attention blocks while preserving the number of oracle calls.
Read the full abstract

Optimization and Control

An Extension of Major-Minor Mean Field Game Theory

This work extends the theory presented in Mean Field Games with a Dominating Player by Bensoussan, Chau and Yam on mean field games with a dominating player, to the case in which the utility and cost functions depend not only on the law of the states, but on the joint state–control law. We incorporate the conditional distribution of the state–control pair of the representative agent given the common noise of the dominating player. In addition, we generalize the role of the dominating player to include the direct impact of its controls $u_0$ on the dynamics and functionals of the system. The optimization problems are reformulated in terms of the conditional distribution of the state–control pair, the necessary optimality conditions are established via stochastic maximum principles, and a coupled SHJB–FP system of equations is obtained that synthesizes the equilibrium conditions. This framework provides a significant extension of the existing literature on MFG with a dominating player.
Read the full abstract

Mathematical Physics

A symplectic geometric origin of universal quartic modified dispersion relations

We show that quartic modifications of relativistic dispersion relations arise generically from deformation-quantized phase spaces under minimal kinematical assumptions relevant to quantum gravity. When the kinematics admits an integral symplectic structure, a compatible almost-complex structure, and a gauge-invariant two-form sector, the leading Planck-scale correction is controlled by a single geometric length scale. We establish this result through three independent approaches: Fedosov-Berezin quantization, spectral geometry, and a topos-theoretic formulation, all of which yield the same quartic correction and clarify the origin of its apparent universality.
Read the full abstract

Optimization and Control

Liquidity Pools as Mean Field Games with Transaction Costs

This paper extends the theoretical framework introduced in Liquidity Pools as Mean Field Games: A New Framework, where the interactions among traders in a constant product market-making protocol were modeled using mean field games (MFG). In this extension, transaction costs are incorporated into the traders' inventory dynamics, modeling the impact of pool fees on trading decisions. Traders operate at a mid-price adjusted for transaction costs, introducing a new dynamic for the DAI inventory. The existence of MFG solutions for this new trader game is established, taking these additional costs into account. The traders' optimization strategies and the conditions for equilibrium existence are discussed, providing a solid foundation for future research.
Read the full abstract

Number Theory

Quantitative bounds on integrality for post-critically finite maps

Let $K$ be a number field with algebraic closure $\overline{K}$ and let $S$ be a finite set of places of $K$ that contain all the archimedean places. For an integer $d \ge 2$, consider the unicritical polynomial family $f_{d,c}(z) = z^d + c$. Recently, Benedetto and Ih studied the distribution of post-critically finite parameters $c$ that are $S$-integral relative to a fixed point $\alpha \in \overline{K}$ such that $f_{d, \alpha}$ is not post-critically finite. In this paper, we study the quantitative aspects of their result. In particular, under some additional assumptions we establish quantitative bounds on the number of $S$-integral post-critically finite parameters in the generalized Mandelbrot set $\mathcal{M}_{d, v}$ relative to a non post-critically finite parameter $\alpha$ as $\alpha$ varies over number fields of bounded degree.
Read the full abstract

Mathematical Physics · Cross-listed from nlin.SI

An Extended Modified Kadomtsov-Petviashvili Equation: Ermakov-Painlevé II Symmetry Reduction with Moving Boundary Application

Here, a novel 2+1-dimensional nonlinear evolution equation with temporal modulation is introduced which admits integrable Ermakov-Painlevé II symmetry reduction. Application is made to obtain exact solution to a class of Stefan-type moving boundary problems for this 2+1-dimensional nonlinear evolution equation. Involutory transformations with origin in autonomisation of certain Ermakov-type coupled systems are extended to 2+1-dimensions and applied to derive a wide 2+1-dimensional class with temporal modulation and which inherits the property of admittance of such hybrid Ermakov-Painlevé II symmetry reduction applicable to certain moving boundary problems.
Read the full abstract

Number Theory

Quantum ergodicity of Eisenstein series for Bianchi groups

We prove the quantum ergodicity of Eisenstein series on the arithmetic hyperbolic 3-manifold $\operatorname{PSL}_2(\mathcal{O}_F)\backslash \mathbb{H}^3$, where $F$ is an imaginary quadratic field with ring of integers $\mathcal{O}_F$ and class number $h_F\geq 1$. This extends the work of Koyama, who proved the result in the case $h_F=1$, and establishes the first instance of quantum ergodicity of Eisenstein series over number fields with nontrivial class groups.
Read the full abstract

Numerical Analysis

Neural network parametrized level sets for image segmentation

Chan-Vese algorithms have proven to be a first-class method for image segmentation. Early implementations used level set methods with a pixelwise representation of the level set function. Later, parametrized level set approximations, such as splines, have been studied and computationally developed to improve efficiency. In this paper, we use neural networks as parametrized approximations of level set functions for implementing the Chan-Vese methods. We show that this approach is efficient because of the equivalence between two layer neural networks and polygonal approximations of level set-based segmentations. In turn, this allows the two-layer network architecture to be interpreted as an ansatz function for the approximate minimization of Chan-Vese functionals. Based on these theory, we extend the classical Chan-Vese algorithm to a data-driven setting, where prior parameters of the network are obtained through unsupervised training on representative image data. These learned parameters encode geometric structures of the data, leading to improved initialization and faster convergence of the Chan-Vese image segmentation.
Read the full abstract

Representation Theory

Classification of degenerate Verma modules over $E(4,4)$

In this paper we classify degenerate Verma modules over the linearly compact Lie superalgebra $E(4,4)$. This completes the description of Verma modules over the exceptional linearly compact Lie superalgebras. As in the other cases all degenerate modules and morphisms between them give rise to infinite bilateral complexes which may be viewed as a generalization of de Rham complexes.
Read the full abstract

Optimization and Control

Clash of MINLP Relaxations: Piecewise Linear vs. Global Parabolic

Solving mixed-integer nonlinear programs (MINLPs) typically relies on constructing relaxations that are easier to tackle than the original problem. Recently, global parabolic (PARA) relaxations were introduced, featuring separable quadratic functions – paraboloids – as global under- or overestimators of general nonlinear constraint functions. So far, the paraboloids are all computed at once by solving a mixed-integer linear program (MIP). For small tolerances or wide function domains, the corresponding MIP grows in size and is eventually intractable, reventing a meaningful comparison with established relaxation techniques. We therefore propose a novel iterative method to compute PARA approximations that succeeds on all tolerance-domain combinations where the original one has failed. The computational study is preceded by a thorough theoretical explanation and analysis. Finally, the improved method enables a computational comparison with piecewise linear (PWL) relaxations in terms of runtime on general MINLP instances. The results show that the modern solver SCIP can solve PWL relaxations faster when the tolerance is high, shifting strongly in favor of PARA for tighter tolerances. We attribute the effect to the difference in the corresponding problem size: PWL relaxations introduce binary variables to identify the active linear piece and their number grows with decreasing tolerance. PARA, on the other hand, does not require additional variables such that the dimension is maintained. For problems with at least one (co)sine constraint, the effect significantly amplifies. Thereby, for medium tolerances, PARA relaxations outperform SCIP stand-alone. Applied problems like alternating current optimal power flow (AC-OPF) feature such constraint types, leaving PARA a viable relaxation strategy.
Read the full abstract

Differential Geometry

Pinching rigidity theorems for normal scalar curvature

Let $M^n$ be an $n$-dimensional closed minimal submanifold immersed in the unit sphere $\mathbb{S}^{n+m}$. Denote by $S$ and $\rho^{\perp}$ the squared norm of the second fundamental form and the normal scalar curvature of $M^n$, respectively. Let $\{A^{\alpha}\}_{\alpha=n+1}^{n+m}$ be the shape operators of $M^n$ with respect to a local orthonormal normal frame. Denote by $\lambda_{1}$ the largest eigenvalue of the positive semi-definite symmetric matrix $\mathcal{A}=(\langle A^{\alpha},A^{\beta}\rangle)_{m\times m}$. We show that if $\lambda_{1}\leqslant n$ and $\rho^{\perp}\leqslant \left[{\sqrt{2}n(n-1)}\right]^{-1} \mathop{\inf}\limits_{p\in M}(n-\lambda_{1})(p)$, then $\rho^{\perp}\equiv 0$, which means the normal bundle of $M^n$ is flat, and further we give the classification of $M^n$.
Read the full abstract

Group Theory

On the number of non-cyclic subgroups of finite p-groups

Let $G$ be a finite $p$-group and $\delta(G)$ denote the number of all non-cyclic subgroups of $G$. In this paper, an upper bound for $\delta(G)$ is obtained. Furthermore, we prove that $\delta(G)\leq \delta(M_p(1, 1, 1) \times C_{p}^{n-3})$ (if $p=2$, then $\delta(G)\leq \delta(D_8\times C_{2}^{n-3})$), for any non-elementary abelian $p$-group $G$ of order $p^n$.
Read the full abstract

Commutative Algebra

Annhilators of local cohomology modules over modular invariant rings and Dickson polynomials

Let $\mathbb{F}_q$ be a finite field with $q = p^s$ elements. Let $V$ be a $d$ dimensional vector space over $\mathbb{F}_q$ and let $G$ be a subgroup of $GL(V)$. Let $R = \mathbb{F}_q[V] = \text{Sym}_{\mathbb{F}_q}(V^*)$ and let $G$ act naturally on $R$. Set $S = R^G$. Let $\mathbf{d}_{d,0}, \mathbf{d}_{d, 1}, \ldots, \mathbf{d}_{d, d-1} \in S$ be the Dickson polynomials with $\deg \mathbf{d}_{d,i} = q^d - q^i$. Let $I$ be a homogeneous ideal of $S$ and let $H^i_I(S)$ be the $i^{th}$-local cohomology module of $S$ with respect to $I$. Let $J_i = \sqrt{\text{ann} H^i_I(S)}$. Assume $J_i \neq 0$ and $\dim S/J_i = d - g$. Then we show that $\mathbf{d}_{d,0}, \ldots, \mathbf{d}_{d, d - g + 1} \in J_i$. We give several applications of our results. An application is a considerably simpler proof of Landweber-Stong conjecture.
Read the full abstract

Metric Geometry

A Complete Graphic Statics for Rigid-Jointed 3D Frames. Part 3: Loops for Kinematics

In Part 3 of this sequence of papers, the kinematic behaviour of 3D frame structures is described using the loop formalism that was developed in Part 2 to describe equilibrium. There, the notions of polygons, polyhedra and polytopes that form the geometric toolbox underlying graphic statics were replaced by the more general concept of CW-complexes from algebraic homology. The six components of the stress resultant acting on any cut face of a bar in a rigidly-jointed framework were represented by the oriented bivector areas of the six projections of a loop in a 4D-space, with three components representing the force and three components representing the moment. In this paper, projected areas of loops in 4D will represent kinematic variables, with three projected areas representing the displacement of a point on the frame, and three other projected areas representing the rotation of the structure at that point. The 4D setting for the theory consists of the usual three dimensions of physical space together with a fourth dimension for the stress function. Virtual Work then manifests as a top form (an oriented 4-volume) in this 4D setting, being the integral over the structure of the wedge product of bivectors representing the local equilibrium and kinematic variables.
Read the full abstract

Optimization and Control

Mapping Austria's Natural Gas and Hydrogen Infrastructure Plans

This paper presents a comprehensive, spatially disaggregated dataset of Austria's natural gas and hydrogen infrastructure towards 2040. The dataset covers the complete gas transmission and distribution networks down to the medium-pressure level and integrates hydrogen expansion plans from the Austrian Gas Grid Management. Transmission infrastructure is reconstructed from ENTSOG maps, converted into a topologically consistent graph representation, and enriched with technical attributes through automated spatial matching with open-source datasets such as OpenStreetMap and Global Energy Monitor. Distribution networks and infrastructure modifications are implemented using QGas, a newly developed GIS-based tool for graph-based infrastructure manipulation. To enable forward-looking energy system analyses, the dataset explicitly represents the stage-wise transition from natural gas to hydrogen infrastructure within a single dataset. Repurposed and newly constructed hydrogen pipelines are integrated within a unified network topology using node splitting and time dependent connector elements, enabling consistent modeling of parallel natural gas and hydrogen operation over time. The resulting dataset provides a detailed representation of Austria's gas and hydrogen infrastructure, including 586 natural gas pipeline segments (5000 km), 113 repurposed segments (1250 km), and 39 newly constructed hydrogen segments (820 km), connecting 720 nodes. Moreover, it includes a comprehensive set of gas demands, biogas production facilities, storage units, electrolyzers, and compressor elements, making it directly applicable for energy system optimization models.
Read the full abstract

All authors Marco Quantschnig, Yannick Werner, Thomas Klatzer, Sonja Wogrin

Optimization and Control · Cross-listed from cs.LG

Optimal uncertainty bounds for multivariate kernel regression under bounded noise: A Gaussian process-based dual function

Non-conservative uncertainty bounds are essential for making reliable predictions about latent functions from noisy data, and thus, a key enabler for safe learning-based control. In this domain, kernel methods such as Gaussian process regression are established techniques, thanks to their inherent uncertainty quantification mechanism. Still, existing bounds either pose strong assumptions on the underlying noise distribution, are conservative, do not directly apply in the multi-output case, or are difficult to integrate into downstream tasks. This paper addresses these limitations by presenting a tight, deterministic bound for multi-output functions in Reproducing Kernel Hilbert Spaces (RKHSs) subject to bounded noise. It is obtained through an unconstrained, duality-based formulation, which shares the same structure as classic Gaussian process confidence bounds, and can thus be straightforwardly integrated into downstream optimization pipelines. We show that the proposed bound generalizes existing results and illustrate its application using an example inspired by quadrotor dynamics learning.
Read the full abstract

All authors Amon Lahr, Anna Scampicchio, Johannes Köhler, Melanie N. Zeilinger

Probability

Pattern block method for generating random numbers : Reformulation and generalization of the Ziggurat method using conditional random variables

The Ziggurat method is an efficient rejection sampling technique for generating one-dimensional normally distributed random numbers. This study proposes the pattern block method, a generalization of the Ziggurat method. The pattern block method enables the generation of random numbers from multimodal density functions and multidimensional distributions. The effectiveness of the pattern block method is demonstrated through several examples.
Read the full abstract

Dynamical Systems

Robust transitivity without sectional-hyperbolicity

For any integer $n \geq 5$, we construct an $n$-dimensional $C^1$ vector field exhibiting a robustly transitive singular attractor which is not sectional-hyperbolic. Nevertheless, the attractor is singular-hyperbolic. This provides the first such examples improving some features of the constructions in [17, 32].
Read the full abstract

All authors A. Arbieto, W. Britto, C.A. Morales, E. Rego

Information Theory

Directivity Enhancement of Movable Antenna Arrays with Mutual Coupling

In conventional antenna arrays, mutual coupling between antenna elements is often regarded as detrimental. However, under specific conditions, it can be harnessed to enhance the far-field directivity (i.e., beamforming gain). Theoretically, the directivity of an N-antenna superdirective array over the endfire direction can reach N^{2}, significantly exceeding the directivity of a traditional uncoupled array which is N over all directions. This paper investigates the potential of mutual coupling effects in movable antenna (MA) arrays for directivity enhancement. A low-complexity algorithm called Greedy Search and Gradient Descent (GS-GD) is proposed to optimize the antenna positions for maximizing the array directivity over any given direction, where the antenna positions are first selected sequentially from discrete grid points and then continuously refined through gradient descent (GD) optimization. Numerical results demonstrate that the optimized MA array design by exploiting the antenna coupling achieves significant directivity gains compared to the conventional uniform linear array (ULA) without antenna coupling over all directions. Additionally, the proposed GS-GD algorithm is shown to approach the global optimum closely in most directions.
Read the full abstract

All authors Wei Xu, Lipeng Zhu, Wenyan Ma, An Liu, Rui Zhang