Peer Revenue Past citations are no guarantee of future returns

Papers

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

Mathematics

math Watch Mathematics (sign in to use)

Tuesday 19 May

Optimization and Control · Cross-listed from cs.RO

Aerial Inspection Behaviors via RL-based Quadrotor Control for Under-canopy Forest Environments

This paper addresses the problem of using a deep Reinforcement Learning (RL)-based low-level Quadrotor controller within an autonomous Quadrotor navigation stack for aerial inspection missions in under-canopy forest environments. Specifically, the article presents an end-to-end (mapping states to RPMs) Quadrotor control policy that achieves inspection view-pose tracking (simultaneous position and yaw reference tracking), which is crucial for various target inspection behaviors and point-to-point navigation in forests. To ensure safe and reliable deployment of the end-to-end RL controller in long-range missions, this article utilizes a higher navigation guidance layer comprising of a Traveling Salesman Problem planner (TSP) and a Rapidly-exploring Random Tree Star (RRT*) planner. Over a known map of a forest and a set of user-specified inspection regions, the TSP planner finds the optimal visitation sequence. Between two target regions, collision-free paths that respect the tracking limitations of the lower end-to-end RL policy are generated by an RRT* planner. Through five target inspection scenarios, this article demonstrates that an RL-based motor-level stabilizing controller, supported by a navigation guidance layer, can be used effectively as the low-level inspection execution module for under-canopy forest inspection missions.
Read the full abstract

All authors Fausto Mauricio Lagos Suarez, Akshit Saradagi, Vidya Sumathy, Viswa Narayanan Sankaranarayanan, George Nikolakopoulos

Monday 18 May

Numerical Analysis · Cross-listed from physics.flu-dyn

Viability of Tensor Train Methods for Geophysical Fluid Dynamics

Tensor train (TT) methods have recently gained popularity for accelerating the solving of systems of PDEs. Here, we evaluate the performance of TT methods in the context of geophysical fluid dynamics (GFD) using the shallow water equations and a discretization scheme employed by the ocean component of the Energy Exascale Earth System Model (E3SM). Through a suite of four test cases of increasing complexity, we evaluate TT methods in terms of how much TT is able to compress the model state, the error incurred by the TT approximation, and the speedup obtained by TT versus an optimal standard non-TT implementation in a representative subproblem. We show that though TT is able to effectively compress and speed up simple flows, it struggles to efficiently represent more complex states that are common in realistic GFD applications.
Read the full abstract

Complex Variables

Prabhakar function and unified fractional kinetic equation in bicomplex space

The Mittag-Leffler type functions arise naturally in the solution of fractional order integral and differential equations, especially in the investigations of the fractional generalization of the kinetic equation. This article introduces a bicomplex extension of the Prabhakar function, a generalization of the Mittag-Leffler function commonly used in fractional calculus. We explore the analyticity and determine the region of convergence for this new bicomplex Prabhakar function. Several fundamental properties are established, including its integral representations, recurrence formulas, and differential relations. Furthermore, we compute the bicomplex Laplace and Mellin transforms of the function, which are useful for solving differential and integral equations. Finally, we analyze a fractional kinetic equation where the bicomplex Prabhakar function appears both in the equation and in its solution, demonstrating its applicability in complex systems involving fractional dynamics.
Read the full abstract

Number Theory

Port Fillings for Primary Pseudoperfect Numbers

Erdős asked whether there are infinitely many finite sets of distinct primes $p_1<\cdots<p_k$ and positive integers $m$ such that \begin{equation}\label{eq:erdos-original} \frac1{p_1}+\cdots+\frac1{p_k}=1-\frac1m. \end{equation} This is Erdős Problems \#313~\cite{ErdosProblems313}. As recalled below, it is equivalent to the infinitude of primary pseudoperfect numbers. Following Butske, Jaje, and Mayernik~\cite{ButskeJajeMayernik}, a squarefree positive integer $n$ is a primary pseudoperfect number if \begin{equation}\label{eq:ppn-def} \frac1n+\sum_{p\mid n}\frac1p=1, \end{equation} where the sum is over the prime divisors of $n$. OEIS A054377~\cite{OEISA054377} records the initial values \[ \begin{array}{c} 2,\ 6,\ 42,\ 1806,\ 47058,\\[2pt] 2214502422,\ 52495396602. \end{array} \] and the eight-prime-factor example \[ \text{\seqsplit{8490421583559688410706771261086}}. \] Butske, Jaje, and Mayernik proved by computation that for each $r\le 8$ there is exactly one primary pseudoperfect number with $r$ distinct prime factors~\cite{ButskeJajeMayernik}. This result gives a useful baseline, but it does not address later layers or the infinitude problem. This paper uses a local language for residual equations. A port is a pair $(R,c)$, and a squarefree integer $B$ fills it if \[ \Delta_{R,c}(B):=cB-R\partial(B)=1. \] The corresponding reciprocal form is \[ \sum_{q\mid B}\frac1q+\frac1{RB}=\frac cR. \] The product rule for the arithmetic derivative gives the composition law for ports. This law separates fillings inherited from smaller primary pseudoperfect numbers from fillings that are primitive relative to the fixed residual equation. The unconditional results of the paper are as follows.
Read the full abstract

General Mathematics

Bitwise Triangular Coordinates for Central Products of Quaternion Groups: Floretion Base Vectors, Digitwise S3-Actions, and Centralizer Tiles

We give a self-contained treatment of the floretion coordinate model for the central product of $n$ copies of the quaternion group $Q_8$. Positive basis elements are words of length $n$ in the alphabet $\{1,2,4,7\}$, identified with $i,j,k,e$, and the resulting real algebra is $\mathbb{H}^{\otimes n}$. In these coordinates, Boolean multiplication, recursive triangular tilings, digitwise $S_3$-actions, reflection anti-automorphisms, centralizer tiles, and axis-landing phenomena admit a common description. A local XNOR/AND rule recovers quaternionic basis multiplication in every order. The centroid map for the recursive triangular tiling intertwines the digitwise $S_3$-action with the dihedral action on the triangle, while odd digit permutations reverse multiplication order. Synchronized cyclic changes in selected coordinates produce equilateral centroid triangles; their oriented vertex product gives an explicit symbolic center, and a parity criterion characterizes when this agrees with the Euclidean center. For every non-unit basis word, the centralizer in the signed group has cardinality $4^n$ and its positive tile set occupies one half of the order-$n$ tiling. Its positive-product component $C_+(b)$ is invariant under the global cyclic digit action and, apart from the identity tile, decomposes into centered equilateral $C_3$-orbits; the negative-product component $C_-(b)$ is a transversal of the complementary orbit family. A parity-dependent family from the signed centralizer decomposition further shows that products symmetric about one triangular axis may land on another axis, lose all triangular reflection symmetry in odd orders, or acquire enhanced symmetry in order two.
Read the full abstract

General Mathematics

Distributions of Inversions and Descents over Integer Compositions

We derive a generating function for the number of integer compositions of $n$ into $k$ parts (i.e., $k$-compositions of $n$) with a given number of inversions, and obtain similar results for $k$-compositions of $n$ with a given number of descents. Our approach relies on a known bijection that associates each integer composition $\sigma$ with a pair $(\pi,\lambda)$, where $\pi$ is a permutation and $\lambda$ is an integer partition. We show that the distribution of inversions and the distribution of descents over $k$-compositions are related, respectively, to the distribution of (maj,inv) and to the distribution of (inv,des) over permutations of $\{1,2,\ldots,k\}$, where maj, inv, and des denote the classical permutation statistics major index, inversion number, and descent number, respectively.
Read the full abstract

General Mathematics

Vector Invariance and Structural Closure of Julia-Type Iterations in Clifford Algebra

In this paper, we introduce a Clifford algebra framework for Julia-type dynamics driven by the geometric product. The nonlinear iteration \[ f(\vec{x}) = (\vec{x}\diamond \vec{n})^p \diamond \vec{n} + \vec{c}, \qquad p \ge 2, \] is studied in a real $n$-dimensional inner-product space $V$, where $\vec{x}, \vec{n}, \vec{c} \in V$ and $\vec{n}$ is a unit vector. The main result reveals a previously unreported invariance phenomenon: although the geometric product generates higher-grade multivector components at intermediate stages, a built-in grade-reduction mechanism ensures complete collapse back to the vector subspace. Consequently, the Clifford Julia operator is shown to be closed on $V$, and the iteration defines a well-posed nonlinear dynamical system in arbitrary dimensions. This invariance is established through a structural decomposition of the Clifford product and an inductive closure argument, supported by explicit verification in low-dimensional cases and a general proof in $\mathbb{R}^n$. The results demonstrate that classical Julia dynamics can be consistently extended beyond the complex plane into higher-dimensional geometric algebra without loss of geometric interpretability. The framework opens a new direction for fractal-type dynamics in Clifford algebras, providing a unified algebraic setting for higher-dimensional invariant-preserving iterative systems.
Read the full abstract

History and Overview of Mathematics

God numbers for Graphs, Games and Groups

We describe and axiomatize finite solitaire puzzles and zero sum sequential games graph theoretically. Zermelo's theorem telling that there is a win for one of the players or a draw follows from the definitions. The god number is a geometric quantity that quantifies the number of moves necessary to solve the puzzle. In the solitaire case, the god number is the minimal distance from the initial state $v$ to the solution space $A$. If $v$ and $A$ are not specified, the god number is the graph diameter. God number computations are related to combinatorial sorting problems and is a NP-complete problem in general even when restricted to concrete sliding problems. In the two-player case, the god number is a minimax critical value: it minimizes the maximal game event length over the set of all strategies. A strategy is a sub-graph of the game graph that contains the initial vertex. The definition is done so that a “mate in k" chess problem has god number k. As for examples: in the solitaire case, we look at group games like Rubik type problems, transposition games related to sorting, at sliding puzzles like the 15 puzzle or rainbow ball, or the tower of Hanoi. For two-player games, we illustrate the story using examples of small chess games, a small card game or tic-tac-toe type problems.
Read the full abstract

All authors Z. Adams, M.Z. Cassim, C. Hou, O. Knill, V. Seco Roopnaraine, M.H. Saleem

Information Theory · Cross-listed from cond-mat.stat-mech

The Thermodynamic Costs of Simple Linear Regression

The construction of models from data is a significant contributor to the energetic costs of computation. Because of this, understanding how foundational thermodynamic bounds apply to modeling algorithms will be increasingly important. Here, we study the thermodynamic costs of a basic and fundamental modeling algorithm: simple linear regression. Following Landauer, we approximate the thermodynamic lower bound on irreversibly performing both exact linear regression and linear regression via stochastic gradient descent as implemented on floating-point numbers. From this, we derive energycost aware scaling laws for the optimal dataset size for training a linear regression model given a generalization error dependent demand for inference. Additionally, we discuss a method to lower bound the entropy production from the mismatch cost for algorithms with continuous input variables.
Read the full abstract

All authors Samuel H. D'Ambrosia, Sultan M. Daniels, Michael R. DeWeese, Anant Sahai

Statistics Theory

Inference Functionals and Observation Operators for Distributional Statistical Models

This paper generalises inference functions (Godambe, 1960) to distributional statistical models, in which each probability measure is represented by a distribution–kernel pair $(T_\theta, \varphi) \in \mathcal S'(\mathbb R) \times \mathcal S(\mathbb R)$. The generalisation is strategically motivated: the key properties of maximum likelihood estimation-consistency and asymptotic normality -derive not from maximising the likelihood but from the MLE being the root of a regular inference function. Extending inference functions to the distributional setting provides an optimality theory for models lacking classical densities or finite moments. The extension requires enlarging the notion of observation. We introduce observation operators $\mathcal O : \mathcal S'(\mathbb R) \to \mathcal Y$ mapping distributional models to an observation space, and define inference functionals as estimating equations composed with these operators. The framework encompasses classical point observations, interval-censored data, convolutional measurements, and transform-based statistics. We establish asymptotic theory (consistency, asymptotic normality, Godambe optimality) under mild conditions and derive a hierarchy of information bounds – classical Fisher information dominates the information available through the observation operator, which in turn dominates the information captured by any inference functional – via the Hájek–Le~Cam convolution theorem. The two gaps quantify distinct sources of information loss: the observation mechanism and the choice of inference functional. Examples include sinusoidal inference functions for heavy-tailed distributions, interval-censored location inference, elliptically contoured models, and nuisance parameters via the Bhapkar–Godambe projection.
Read the full abstract

Number Theory

A formula of counting divisors in integers rings: a generalization of the divisor function $d_0(n)$

In this paper we establish a formal connection between the structure of ideals in integers rings and the theory of additive combinatorics. For integers rings with cyclic class groups, we prove a structural theorem demonstrating that every non-zero ideal can be decomposed into a maximal principal part and a product of ideals whose total length is bounded by the Davenport constant. With this decomposition we find divisors for generators of the ideal $I=(\alpha, \beta)$. The central result of this work is the derivation of a closed formula using character theory over finite abelian groups to count the exact number of zero-sum subsequences of a given sequence. Under the established correspondence between principal ideals and zero-sum sequences, this formula provides a precise counting of the principal ideal divisors of any given ideal, and therefore counting common divisors of generators of the ideal $I=(\alpha,\beta)$. This result constitutes a natural generalization of the classical divisor function $d_0(n)$ from unique factorization domains to any Dedekind domain with a finite class group. Finally, we characterize irreducible elements in $\mathcal{O}_K$ based on the counting of these zero-sum subsequences.
Read the full abstract

Representation Theory

On geometrically reductive tensor categories

We prove the conjecture that higher Verlinde categories are geometrically reductive. This is one of the two properties required in order for recent results on algebraic geometry in tensor categories to apply to these categories. We also reduce two further conjectures concerning geometric reductivity for tensor categories to other conjectures appearing in the literature.
Read the full abstract

Optimization and Control · Cross-listed from cs.RO

A Heuristic Approach for Performance Tuning in RL-based Quadrotor Control via Reward Design and Termination Conditions

Reinforcement learning (RL)-based quadrotor control policies have achieved impressive performance in tasks such as fast navigation in cluttered environments and drone racing, where the focus is on speed and agility. However, in several applications, such as infrastructure inspection, it is critical to achieve precise, controlled maneuvers with tunable performance. In this article, we present a novel heuristic approach to achieve tunable performance in RL-based Quadrotor control through reward design and termination conditions. We present a novel reward structure containing dual bandwidth exponentials that achieves a baseline critically damped response in setpoint tracking, with low steady-state errors. When trained with a Proximal Policy Optimization (PPO) algorithm, in conjunction with episode truncation conditions, the desired performance is achieved in 6 million time steps in a sample-efficient manner. In order to tune the performance about the baseline behavior, we present intuitive heuristic rules to adjust the reward weights and exponential coefficients to achieve faster (acrobatic-like) and slower (inspection-like) settling time performance, while retaining the baseline critically damped response and approximately 2% steady-state error. We evaluate the three RL policies (baseline, acrobatic, and inspection) across 100 trials and show accurate and tunable performance in position and yaw tracking from random initial conditions, thereby demonstrating the effectiveness of the proposed heuristic approach.
Read the full abstract

All authors Fausto Mauricio Lagos Suarez, Akshit Saradagi, Vidya Sumathy, George Nikolakopoulos

Number Theory

On nonconvex constellations among primes II: (458,3240)

Extending our work on the $k$-tuple conjecture, we previously applied those methods to the Engelsma counterexamples (narrow constellations) of length $J=459$ and span $|s|=3242$. Here we extend that analysis to the $116$ Engelsma counterexamples of length $J=458$ and $|s|=3240$. We track the evolution of these $116$ counterexamples from inadmissible driving terms starting in the cycle of gaps ${\mathcal G}(11^\#)$ up through their first appearance in ${\mathcal G}(113^\#)$. We continue developing primorial coordinates for each admissible instance through a breadth-first exhaustive search through ${\mathcal G}(211^\#)$. Each of the $(458,3240)$ constellations sits inside a $(459,3242)$ constellation, which we call its parent. We show that no $(458,3240)$ constellation occurs outside of its parent until the cycle ${\mathcal G}(227^\#)$. The early evolution of the $(458,3240)$ constellations is dominated by the evolution of their parents, which we have previously studied. For each $(458,3240)$-counterexample we calculate its asymptotic relative population, among other constellations of length $J=458$.
Read the full abstract

Statistics Theory · Cross-listed from stat.ME

Weak Convergence and Gaussian Limits for a General-Dimensional Origin-Invariant Cramér–von Mises Statistic

We introduce a general-dimensional origin-invariant Cramér–von Mises statistic $\bar{\omega}_d^2$ for testing complete spatial randomness, defined by averaging corner-oriented empirical-distribution-function discrepancies over all $2^d$ corners of $[0,1]^d$. The statistic has a closed-form $O(n^2)$ computing formula. For $d=1$ it reduces to the classical rank-based Cramér–von Mises statistic, while for $d=2$ its computing formula agrees with Zimmerman's origin-invariant statistic. Under iid complete spatial randomness, the associated empirical process converges in $\ell^\infty$ to a centered Gaussian process with an explicit cross-corner covariance kernel. The continuous mapping theorem yields a quadratic Gaussian limit governed by a positive trace-class covariance operator with trace $2^{-d}-3^{-d}$. We also connect fixed-count complete spatial randomness with the homogeneous Poisson formulation. For strictly stationary alpha-mixing sequences with uniform marginals, we establish covariance summability, the long-run variance limit, and a finite-dimensional all-corners Gaussian limit. Monte Carlo experiments illustrate null calibration and sensitivity to selected alternatives. The statistic is consistent against fixed iid alternatives whose distribution functions differ from uniformity on a set of positive measure.
Read the full abstract

Commutative Algebra

An algorithm for invariants of elementary abelian groups

When we consider a finite abelian group acting linearly on a polynomial ring, we can find monomial generators for the subring of invariants. By Noether's degree bound and Hilbert's finiteness theorem, we know that there are finitely many minimal generators, but efficiently finding a generating set is not a trivial task. We present a new algorithm for computing the invariant ring for elementary abelian groups acting on polynomial rings with complex coefficients (or any other field of characteristic zero). We follow a two-step process: first we generate a collection of $n-k$ "seed" invariants by calculating the kernel of a weight matrix that encodes our action. After we find the seeds, we "grow" them into a generating set for the invariant ring by exploiting the lattice structure of invariants modulo $p$. Our algorithm performs better than the one currently available in Macaulay2, allowing us to compute invariants more quickly in this setting.
Read the full abstract

All authors Sasha Arasha, Marcus Cassell, Mal Dolorfino, Francesca Gandini, Gordie Novak, Daniel Qin, Sumner Strom

Information Theory · Cross-listed from stat.ML

Information Processing Capacity of Stationary Physical Systems: Theory, Data-efficient Estimation Methods, and Photonic Demonstration

Physical computing systems provide a promising route toward hardware-native machine learning, but their computational capabilities remain difficult to characterize in a principled, task-independent, and data-efficient way. We extend the Information Processing Capacity (IPC) framework to stationary physical computing systems and establish several fundamental results: individual capacities are bounded between zero and one, their sum over a complete basis is bounded by the number of readouts, and noise strictly reduces this bound. We address the finite-sample estimation of IPC and derive the asymptotic form of the systematic positive bias affecting naive estimators. Building on these results, we introduce data-efficient estimation methods based on Richardson extrapolation and Sobol quasi-random sampling. We validate the framework experimentally using a photonic computing system based on picosecond laser pulses propagating through a nonlinear optical fibre. By varying the laser power and fibre length, we observe systematic shifts of the IPC distribution toward higher-order nonlinear capacities induced by the Kerr effect. Finally, we demonstrate that the total IPC strongly correlates with performance on benchmark machine-learning tasks and provides a reliable estimate of the effective dimensionality of the system. These results establish IPC as a practical bridge between the intrinsic dynamics of physical computing systems and their machine-learning performance.
Read the full abstract

Information Theory

Correcting Tail Deletions in Rank Modulated Composite Encoding for Data Storage in DNA

We study the combination of two recent coding approaches, in the context of DNA based data storage. Composite DNA alphabets leverage properties of the DNA synthesis and sequencing process. A composite symbol does not represent a single nucleotide, but rather a designed mixture of DNA nucleotides. Using the high multiplicity that is intrinsic to synthesis and sequencing a composite symbol consists of frequencies in the mixture. Rank modulation codes use permutations to represent information. Combining the two, we construct encoding that uses permutations of nucleotide frequencies rather than the exact frequency values. Codes for this approach were addressed in previous work, under Kendall's tau distances. In this work we study deletion and insertion codes. We present bounds and constructions of efficient codes defined over partial permutations.
Read the full abstract

Mathematical Physics · Cross-listed from gr-qc

Weak cosmic censorship for the circularly symmetric Einstein-scalar field system in $2+1$ dimensions

We prove the weak cosmic censorship conjecture in $2+1$ spacetime dimensions for the circularly symmetric Einstein-scalar field system in the presence of a negative cosmological constant $\Lambda<0$. More precisely, we show that for any integer $k\geq2$, the maximal development of generic $C^k$ initial data does not contain naked singularities. An essential step of the proof is establishing the presence of a mass gap. In particular, this implies that all naked singularities have infinite blueshift, which represents the main instability mechanism.
Read the full abstract

Analysis of PDEs

Solutions to Monge-Ampère Equations with Low Codimensional singularities

We construct solutions to Monge-Ampère equations whose Monge-Ampère measures contain singular components supported on low codimensional sets. We also study the regularity of such solutions. To motivate our construction, we present examples arising from optimal transport where the potential of optimal transport maps satisfy Monge-Ampère equations similar to the ones we study.
Read the full abstract

Combinatorics

Limit Laws for Consensus Protocols on the Complete Graph

We study a distributed consensus problem on a complete communication network of $n$ vertices, each holding one of two opinions. The vertices communicate in rounds, possibly in the presence of adversarial noise, and exchange information until they all agree on a single opinion. We consider a general class of protocols, where the vertices randomly sample neighbors and update their own opinion according to an update function $f$ depending on the sampled opinions. A prominent example is the $k$-maj protocol, where every vertex adopts the majority opinion of $k$ randomly sampled neighbors, breaking ties uniformly. We consider the runtime $R_n$ that is the number of rounds until all vertices agree on the same opinion, which we call the dominating opinion $D_n$. In our main result we describe the limiting distributions of these two key quantities for a large class of update functions $f$, for arbitrary initial configurations and under the presence of an adversary who may alter the opinions of up to $o(\sqrt{n})$ vertices in each round. We show that there are $f$-specific constants $\gamma, m > 0$ such that $R_n$ centers around $\mu_n = \frac{1}{2}\log_\gamma n + \log_m\ln n$, and we describe the asymptotic distribution of $R_n - \mu_n$. In particular, we show that it does not converge, and that it becomes asymptotically periodic both in the $\log n$ as well as the $\log\log n$ scale. Applied to $k$-maj, our results show, among other things, that $\gamma_{k\text{-maj}} = \binom{k-1}{\lfloor k/2 \rfloor}2^{1-k}k \sim ({2k}/{\pi})^{1/2}$.
Read the full abstract

History and Overview of Mathematics

A Scaling-Parameter Framework for Perimeter and Area in Self-Similar Planar Fractals

The Koch snowflake is a classical example of a planar curve with infinite perimeter enclosing a finite, positive area. Although such examples are well known individually, classical treatments typically analyze each construction in isolation and classify them by similarity dimension. This paper develops a unified parameter-space representation for a class of self-similar planar constructions, organized by two integers – the number of self-similar pieces $N$ and the inverse linear scale factor $r$ – together with two derived growth ratios $\alpha = N/r$ and $\beta = N/r^2$, governing perimeter and area scaling respectively. The $(N,r)$ parameter space is partitioned into three regimes – $N \le r$, $r < N < r^2$, and $N \ge r^2$ – corresponding to qualitatively distinct asymptotic behaviors of perimeter and area jointly. Within the intermediate regime $r < N < r^2$, a construction-class refinement distinguishes additive constructions (region bounded by the iterated curve), which yield positive finite asymptotic area under a stated non-overlap assumption, from subtractive constructions (iterated set itself), which yield zero asymptotic area. This records a structural non-equivalence inside the same dimension class that is not visible from $D = \log N / \log r$ alone. Four worked examples illustrate the framework – the Sierpinski triangle, Sierpinski carpet, Koch snowflake, and a Koch-style construction on a square invented by the author – and four further constructions are analyzed predictively to demonstrate that diagnostic outputs follow from $(N, r, \text{construction class})$ without re-derivation. The contribution lies in formulation and synthesis: the paper consolidates several classical results into a single diagnostic representation in which, given $(N, r)$ and construction class, the asymptotic behavior of perimeter and area can be inferred directly.
Read the full abstract

2605.19128 · 18 May · 17 pages, 5 figures. Author's accepted manuscript. The Python code listings included as Appendices A-C in the published version are omitted here; the code is openly available at https://github.com/pedromarotta/fractal-framework-demo

Mathematical Physics

On the single field formulation in magnetostatics

We systematically discuss the equivalence of two variational formulations of magnetostatics, in terms of magnetization and magnetic field on the one hand and the single field formulation using only magnetic induction. To demonstrate that this link is stable also when the magnetic laws are coupled with other variational static models, elasticity is included in the models as well. Interestingly, despite the fact that the corresponding magnetoelastic energy densities in the material can be computed via Legendre-Fenchel transform in the magnetic state variables, the two formulations are not linked by standard convex duality on the level of the functionals. In addition, convexity and coercivity of the given functional are neither required for the transformation nor always preserved by it.
Read the full abstract

Number Theory

Halász theorems for Gaussian ideals in sectors and short intervals

We prove a quantitative Halász theorem for multiplicative functions on the nonzero ideals of $\mathbb{Z}[i]$, with bounds controlled by pretentious distance to the Archimedean characters $N^{it}$. We also prove a sectorial analogue: under angular non-pretentiousness, the sum of $f$ over ideals lying in a fixed sector is asymptotically given by the expected proportion of the unrestricted sum. Finally, under angular non-pretentiousness and a non-degeneracy condition on conjugate prime pairs, we prove a sectorial short-interval version of the Halász theorem for annular sectors whose radial thickness tends to infinity. The proof of the sectorial short-interval Halász theorem uses angular Fourier expansion, norm-compression to multiplicative functions on $\mathbb{N}$, and a theorem of Mangerel.
Read the full abstract

Complex Variables

Quasiconformal deformations preserving Hilbert's norm and their applications

This paper focuses on estimating the Taylor coefficients for Hilbert spaces of holomorphic functions on the disk using intrinsic features of univalent functions and of Teichmuller spaces. Estimating these coefficients has a long history but still remains an important problem in many geometric and physical applications of complex analysis. We construct quasiconformal deformations of holomorphic functions preserving their Hilbert norm. Such deformations play a crucial role in this subject. Among their applications, a rather complete solution of an old Hummel-Scheinberg-Zalcman problem is obtained.
Read the full abstract

Dynamical Systems

Some topological properties and bi-Lipschitz equivalence of graph-directed attractors

In this paper, we discuss some topological properties of the graph-directed iterated function system (GDIFS) of injective contractions. Further, we establish the existence of a Lipschitz bijection between two different bi-Lipschitz graph-directed attractors. In general, two different bi-Lipschitz graph-directed attractors defined on the same graph need not be bi-Lipschitz equivalent. However, under suitable conditions, we prove that these graph-directed attractors are bi-Lipschitz equivalent.
Read the full abstract

Numerical Analysis

Sharp analysis of sketched least squares and randomized low-rank approximation

Two widely used randomized algorithms are the sketch-and-solve method for least-squares regression and the randomized SVD for low-rank approximation. These algorithms apply a random embedding to compress a target matrix, and they perform computations on the compressed matrix to save computational cost. This paper asks, what is the optimal random embedding in these algorithms? Also, what is the sharpest possible error bound for the optimal embedding? The paper proves that a random orthonormal matrix is minimax optimal for the sketch-and-solve algorithm while any rotation-invariant embedding is minimax optimal for the randomized SVD. Following these results, the paper obtains the best possible error bounds for sketched least-squares and the randomized SVD. Last, empirical experiments provide evidence of universality phenomena, in which several random embeddings lead to similar accuracy to the optimal embeddings in practice.
Read the full abstract

Combinatorics

A Note on the Asymptotic Least Density of Covering Codes in $[q]^n$

In this short note we revisit the upper bound of the asymptotic least density of covering codes of radius $R$ in $[q]^n$ established by Krivelevich, Sudakov, and Vu. We show that by using a slightly different optimization in their core theorem we can obtain a constant factor improvement to their upper bound.
Read the full abstract

Optimization and Control

Reachability-Augmented Dual Dynamic Programming for Optimal Path Parameterization

Optimal path parameterization (OPP) is a fundamental problem for planning trajectories along a prescribed geometric path under kinodynamic constraints and task-dependent objectives. Among existing OPP works, it remains challenging to simultaneously achieve feasibility guarantees, general-objective optimality certificates, and online-compatible computation in a unified framework, especially for third-order OPP (OPP3) with non-convex constraints. This paper proposes reachability-augmented dual dynamic programming (RDDP), a state-grid-free and objective-aware dynamic programming (DP) framework for OPP. The key idea is to replace the relatively complete recourse assumption used in classical dual DP (DDP) with OPP-specific backward reachable sets, and then generate both value-function cuts and trial trajectories only inside these reachable sets. For convex and non-convex OPP, we prove global optimality and Karush-Kuhn-Tucker convergence of RDDP under OPP-specific conditions, respectively. Efficient instantiations are developed for second-order OPP (OPP2) and OPP3. Experiments show that RDDP achieves objective values comparable to convex-optimization baselines with speedups of approximately 15.3 and 4.9 times over the fastest of 4 tested solvers for OPP2 and OPP3, respectively. RDDP also achieves faster convergence than grid-based DP. Compared with reachability-analysis methods, RDDP retains the reachability mechanism while replacing local maximum-control propagation with value-function-guided control selection, thereby enabling objectives beyond traversal time. In summary, RDDP addresses a key capability gap in OPP by unifying certifiable general-objective optimization, reachability-based feasibility preservation, and online-compatible low-dimensional DP computation in a single framework.
Read the full abstract

All authors Yunan Wang, Jizhou Yan, Chuxiong Hu, Zeyang Li

History and Overview of Mathematics

My Warm, Randomly Recorded, Recollections of Professor Richard Askey

These are my memories of moments with Dick and Liz Askey in Russia, Wisconsin, Arizona, and abroad. Dedicated to the Askey family, these recollections span over 40 years and encompass many dramatic changes in the world. Due to this, it is challenging to entirely separate my personal thoughts and feelings from the factual historical account.
Read the full abstract