Differential Geometry
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Sunday 15 February
We introduce an intrinsic subdifferential on Hadamard manifolds induced by a base-point-dependent Busemann coupling, motivated by the scarcity of nonconstant affine functions on non-Euclidean Hadamard manifolds. For proper functions, it characterizes exactly the primal-dual equality pairs in the Fenchel-Young inequality of Bento, Cruz Neto, and Melo (Appl. Math. Optim. 88:83, 2023). A dual-point–tangent-vector bijection yields Fermat's rule and isometry covariance. We characterize everywhere nonemptiness for finite-valued functions by pointwise-attained \(H_p\)-envelope representations, encompassing the radial models with explicit subdifferentials and the nonsmooth distance from the base point. In hyperbolic space, base-point rigidity shows that, for functions differentiable at the base point, nonemptiness is equivalent to global minimality and vanishing of the corresponding Fenchel-Young gap. Two geodesically convex functions on the Poincaré disk prove strict noninclusions between our construction and fixed-direction Busemann subdifferentials, although both recover the classical Euclidean subdifferential. This separation is explained by coupling orientation. In constant negative curvature, a strictly increasing scalar profile describes the asymmetry and defines an intrinsic measure of nonlinearity. We compute this measure at arbitrary radius and derive a lower bound and two-sided estimates for the biconjugation gap; at unit radius, the exact curvature-dependent bound is attained by deviations of both signs.
Information Theory
Energy-Efficient Over-the-Air Federated Learning via Pinching Antenna Systems
Pinching antennas systems (PASSs) have recently been proposed as a novel flexible-antenna technology. These systems are implemented by attaching low-cost pinching elements to dielectric waveguides. As the direct link is bypassed through waveguides, PASSs can effectively compensate large-scale effects of the wireless channel. This work explores the potential gains of employing PASSs for over-the-air federated learning (OTA-FL). For a PASS-assisted server, we develop a low-complexity algorithmic approach, which jointly tunes the PASS parameters and schedules the mobile devices for minimal energy consumption in OTA-FL. We study the efficiency of the proposed design and compare it against the conventional OTA-FL setting with MIMO server. Numerical experiments demonstrate that using a single-waveguide PASS at the server within a moderately sized area, the required energy for model aggregation is drastically reduced as compared to the case with fully-digital MIMO server. This introduces PASS as a potential technology for energy-efficient distributed learning in next generations of wireless systems.
Read the full abstract
Rings and Algebras
The algebraic and geometric classification of noncommutative Jordan superalgebras
The algebraic and geometric classifications of complex $3$-dimensional noncommutative Jordan superalgebras are given. In particular, we obtain the algebraic and geometric classification of $3$-dimensional Kokoris and standard superalgebras, and, due to one-to-one correspondences between suitable superalgebras, we have classifications for generic Poisson-Jordan and generic Poisson superalgebras. As a byproduct, we have the algebraic and geometric classification of the variety of $3$-dimensional anticommutative superalgebras and its principal subvarieties: Lie, Malcev, binary Lie, Tortkara, anticommutative $\mathfrak{CD}$-, $\mathfrak{s}_4$-, anticommutative terminal superalgebras, anticommutative conservative and anticommutative quasi-conservative $\big($rigid$\big)$ superalgebras.
Read the full abstract
Mathematical Physics · Cross-listed from gr-qc
A synthetic Gannon-Lee incompleteness theorem
We prove the Gannon-Lee incompleteness theorem for globally hyperbolic spacetimes. We assume the synthetic null energy condition of Ketterer and a trappedness condition we call "synthetically asymptotically regular". Our result generalizes this classical result to the weighted case. It also motivates and indicates extensions to low regularity, which are deferred to future work.
Read the full abstract
Logic · Cross-listed from cs.CC
Graph Homomorphisms and Universal Algebra
Constraint satisfaction problems are computational problems that naturally appear in many areas of theoretical computer science. One of the central themes is their computational complexity, and in particular the border between polynomial-time tractability and NP-hardness. In this course we introduce the universal-algebraic approach to study the computational complexity of finite-domain CSPs. The course covers in particular the cyclic terms and bounded width theorems. To keep the presentation accessible, we start the course in the tangible setting of directed graphs and graph homomorphism problems.
Read the full abstract
Combinatorics
On the maximum $\sigma$-irregularity of trees with given order and maximum degree
The $\sigma$-irregularity index of a graph is defined as the sum of squared degree differences over all edges and provides a sensitive measure of structural heterogeneity. In this paper, we study the problem of maximizing $\sigma(T)$ among all trees of fixed order $n$ and prescribed maximum degree $\Delta\ge4$. By expressing the problem in terms of edge–degree multiplicities, we derive a linear programming formulation and analyze its dual. This approach yields sharp upper bounds for $\sigma(T)$ and leads to a detailed description of extremal degree–pair distributions. We show that the extremal problem can be completely resolved for the congruence classes $n\equiv1\pmod{\Delta}$ and $n\equiv0\pmod{\Delta}$. When $n\equiv1\pmod{\Delta}$, the linear program admits an integral optimal solution, and the bound for $\sigma(T)$ is tight. When $n\equiv0\pmod{\Delta}$, the linear relaxation is not attainable by any tree; nevertheless, by introducing a penalty function derived from dual slack variables, we determine the exact maximum value of $\sigma(T)$. In both cases, all extremal trees are characterized explicitly and consist exclusively of vertices of degrees $1$, $2$, and $\Delta$, with edges incident to $\Delta$-vertices playing a dominant role.
Read the full abstract
Analysis of PDEs
Nonlocal logistics and nonlinear productions in an attraction-repulsion chemotaxis model: analysis of the global well-posedness
This paper investigates a {{three-component}} chemotaxis system involving both attraction and repulsion effects, as well as a nonlocal logistic-type source term. Mathematically, if $u=u(x,t)$, $v = v(x,t)$ and $w = w(x,t)$ denote the cell distribution, and the attractive and the repulsive chemical signals, the model is then described by \begin{equation*} \begin{cases} u_t = \Delta u - \chi \nabla \cdot (u \nabla v) + \xi \nabla \cdot (u \nabla w) + a u^\alpha - b u^\alpha \int_\Omega u^\beta, & x \in \Omega, \ t > 0, \tau v_t = \Delta v - v + f(u), & x \in \Omega, \ t > 0, \tau w_t = \Delta w - w + g(u), & x \in \Omega, \ t > 0. \end{cases} \end{equation*} Here, $\Omega \subset \mathbb{R}^n$ ($n \geq 1$) is a bounded smooth domain, $\tau\in\{0,1\}$, $a,b,\alpha,\beta,\chi,\xi>0$, the production functions $f(u)$ and $g(u)$ are assumed to satisfy algebraic growth conditions of order $\ell$ and $\rho$, generalizing prototypes of the form $u^\ell$ and $u^\rho$, $\ell,\rho>0$. The work is devoted to proving the global existence and boundedness of classical solutions under a suitable balance between the signal production exponents $\ell, \rho$ and the nonlocal damping exponents $\alpha, \beta$, for regular enough initial data and zero-flux boundary restrictions. In this regard, two main theorems are established for the cases where the chemical signals satisfy either elliptic ($\tau=0$) or parabolic ($\tau=1$) partial differential equations, highlighting how sufficiently strong nonlocal damping prevents the formation of singularities in time. We extend the results obtained in [Chiyo et al., Appl. Math. Optim. 89:9 (2024)], where the fully parabolic ($\tau=1$) and only attraction version is studied. In our context, we establish well-posedness of the system and the long-time behavior of solutions.
Read the full abstract
Optimization and Control · Cross-listed from cs.RO
Muscle Coactivation in the Sky: Geometry and Pareto Optimality of Energy vs. Aerodynamic Promptness and Multirotors as Variable Stiffness Actuators
In robotics and biomechanics, trading metabolic cost for kinematic readiness is a well-established principle. This paper formalizes this concept for aerial multirotors through the introduction of aerodynamic promptness – a dynamic metric analogous to dynamic manipulability in robotics. By formulating redundancy resolution as a geometric multi-objective optimization along task fibers, we rigorously characterize the topological trade-off between energy consumption and promptness. We demonstrate that this interplay is fundamentally governed by fiber geometry. Cooperative actuation regime yields compact fibers with bounded, compatible Pareto fronts. Conversely, antagonistic actuation regime unlocks unbounded fibers, enabling aerodynamic co-contraction that drives promptness to hardware limits at the expense of flight endurance. We establish a structural isomorphism between aerodynamic co-contraction and biologically inspired variable stiffness actuators, introducing a dynamic “flying muscle” paradigm. Ultimately, this framework transitions multirotor allocation from heuristic energy minimization to principled, geometry-aware Pareto navigation, laying foundational theory for the design and control of highly agile aerial platforms.
Read the full abstract
Differential Geometry
The moduli spaces of presymplectic forms on almost abelian Lie algebras
We obtain necessary and sufficient conditions to determine the existence of presymplectic forms of a given rank on all almost abelian Lie algebras. We also study the moduli space of presymplectic forms (this is the set of all closed 2-forms of a given rank under a certain natural equivalence relation) on almost abelian Lie algebras. Most importantly we show that for any almost abelian Lie algebra its moduli space of symplectic forms is finite. Moreover we show that up to such natural equivalence all symplectic forms are permutations of a canonical 2-form. The important step in the proof is obtaining canonical representatives for a certain congruence of matrices, which is of some interest for matrix theory on its own.
Read the full abstract
Group Theory
Classification of abelian Schur groups I
A finite group $G$ is called a Schur group if every Schur ring over $G$ is schurian, i.e. associated in a natural way with a subgroup of the symmetric group $Sym(G)$ that contains all right translations of $G$. The list of all possible abelian Schur groups was obtained by Evdokimov, Kovács, and Ponomarenko in 2016. In two papers, we complete a classification of abelian Schur groups. In the present paper, we study schurity of several groups from the list. First, we prove that a direct product of the elementary abelian group of order 4 and a cyclic group, whose order is an odd prime power or a product of two distinct odd primes, is a Schur group. Second, we establish nonschurity of some other groups from the list.
Read the full abstract
Combinatorics
On the levels of rational regular orthogonal matrices for generalized cospectral graphs
For an $n$-vertex graph $G$ with adjacency matrix $A$, the walk matrix $W(G)$ of $G$ is the matrix $[e,Ae,\ldots,A^{n-1}e]$, where $e$ is the all-ones vector. Suppose that $W(G)$ is nonsingular and $p$ is an odd prime such that $W(G)$ has rank $n-1$ over the finite field $\mathbb{Z}/p\mathbb{Z}$. Let $H$ be a graph that is generalized cospectral with $G$, and $Q$ be the corresponding rational regular orthogonal matrix satisfying $Q^\mathsf{T} A(G) Q=A(H)$. We prove that \begin{equation*} v_p(\ell(Q))\le \frac{1}{2}v_p (\det W(G)) \end{equation*} where $\ell(Q)$ is the minimum positive integer $k$ such that $kQ$ is an integral matrix, and $v_p(m)$ is the maximum nonnegative integer $s$ such that $p^s$ divides $m$. This significantly improves upon a recent result of Qiu et al. [Discrete Math. 346 (2023) 113177] stating that $v_p(\ell(Q))\le v_p (\det W(G))-1.$
Read the full abstract
Optimization and Control · Cross-listed from cs.LG
Fast Catch-Up, Late Switching: Optimal Batch Size Scheduling via Functional Scaling Laws
Batch size scheduling (BSS) plays a critical role in large-scale deep learning training, influencing both optimization dynamics and computational efficiency. Yet, its theoretical foundations remain poorly understood. In this work, we show that the functional scaling law (FSL) framework introduced in Li et al. (2025a) provides a principled lens for analyzing BSS. Specifically, we characterize the optimal BSS under a fixed data budget and show that its structure depends sharply on task difficulty. For easy tasks, optimal schedules keep increasing batch size throughout. In contrast, for hard tasks, the optimal schedule maintains small batch sizes for most of training and switches to large batches only in a late stage. To explain the emergence of late switching, we uncover a dynamical mechanism – the fast catch-up effect – which also manifests in large language model (LLM) pretraining. After switching from small to large batches, the loss rapidly aligns with the constant large-batch trajectory. Using FSL, we show that this effect stems from rapid forgetting of accumulated gradient noise, with the catch-up speed determined by task difficulty. Crucially, this effect implies that large batches can be safely deferred to late training without sacrificing performance, while substantially reducing data consumption. Finally, extensive LLM pretraining experiments – covering both Dense and MoE architectures with up to 1.1B parameters and 1T tokens – validate our theoretical predictions. Across all settings, late-switch schedules consistently outperform constant-batch and early-switch baselines.
Read the full abstract
Algebraic Geometry
The Künneth Formula Of Fundamental Group Schemes
Let $k$ be a field, $f:X\rightarrow S$ a proper morphism between connected schemes proper over $k$, $x\in X(k)$ lying over $s\in S(k)$, $X_s$ the fibre of $f$ over $s$, $\mathcal{C}_X$, $\mathcal{C}_{S}$, $\mathcal{C}_{X_s}$ Tannakian categories over $X,S,X_s$ respectively, $\pi(\mathcal{C}_X,x)$, $\pi(\mathcal{C}_S,s)$, $\pi(\mathcal{C}_{X_s},x)$ the Tannaka group schemes respectively. We give a unified criterion for the exactness of the homotopy sequence of Tannakian fundamental group schemes $\pi(\mathcal{C}_{X_s},x)\rightarrow \pi(\mathcal{C}_X,x)\rightarrow \pi(\mathcal{C}_S,s)\rightarrow 1$. In particular, we obtain the equivalent conditions for the Künneth formula of fundamental group schemes for the product $X\times_k Y$ of two connected schemes $X$ and $Y$ proper over $k$. As an application, we obtain the Künneth formula of certain fundamental group schemes over any field, such as S, N, EN, F, EF, ét, Eét, Loc, ELoc and uni-fundamental group schemes.
Read the full abstract
Statistics Theory
Kernel Estimation Of Chatterjee's Dependence Coefficient
A dependence coefficient suggested by Chatterjee (2021) has attracted considerable attention in recent years. It takes values between 0 and 1, vanishes if and only if the corresponding random variables $X$ and $Y$ are independent, and equals 1 if and only if $Y$ is almost surely a measurable function of $X$. The coefficient coincides, for distributions with continuous marginals, with a dependence coefficient introduced in Dette, Siburg, and Stoimenov (2013). Several estimators of this measure have been proposed, including the kernel-type estimator of Dette, Siburg, and Stoimenov (2013), Chatterjee's rank correlation, and its nearest-neighbour modification by Lin and Han (2023). In this paper, we investigate the asymptotic properties and local power of the kernel-type estimator. We first establish a non-degenerate central limit theorem under independence: after suitable centring, the estimator is asymptotically normal at the rate $\sqrt{n/h_1}$, where $h_1\to0$ is a bandwidth. We then consider Gaussian-copula alternatives with correlation parameter $\rho_n\to0$. By deriving a second-order expansion of the mean, establishing variance stability and proving a central limit theorem under local alternatives, we obtain an explicit limiting power function and identify $(h_1/n)^{1/4}$ as the detection boundary. Under suitable bandwidth conditions, the detection boundary can attain rates arbitrarily close to $n^{-3/8}$. The exponent $3/8$ exceeds the exponent $1/4$ for the test based on Chatterjee's original rank correlation and the exponent approaching $5/16$ for the modified rank correlation of Lin and Han (2023) within the range of their central limit theorem. The proofs combine refined results for double-indexed permutation statistics with Gaussian rank-score expansions and Hermite-chaos projections. A simulation study illustrates the normal approximation and the resulting power properties.
Read the full abstract
Algebraic Geometry
Geometric realisation of hypergeometric local systems
We prove a realisation theorem for irreducible hypergeometric local systems defined over the rational numbers in terms of families of affine varieties in algebraic tori. The families we consider have been studied extensively in the literature and appear in mirror symmetry. Our result holds unconditionally for families with one-dimensional or even-dimensional fibres. It holds under a monodromy assumption for families with fibres of odd dimension greater than one.
Read the full abstract
Analysis of PDEs
Logarithmic Sobolev, Poincaré and Beckner Inequalities on Hyperbolic Spaces and Riemannian Manifolds
We investigate several functional and geometric inequalities on the hyperbolic space $\mathbb{H}^N$, with a primary emphasis on logarithmic Sobolev inequalities, Poincaré inequalities, and Beckner-type inequalities, all studied within the framework of the AB program. The main analytical tool employed throughout this paper is symmetrization. More precisely, our approach relies on an improved version of the Pólya-Szegö inequality on the hyperbolic space, obtained through a careful comparison of the gradient norms of rearranged functions in the hyperbolic and Euclidean settings. For Beckner-type inequalities, we adopt a semigroup approach based on sharp estimates for the heat semigroup, leading to refined interpolation inequalities between Poincaré and logarithmic Sobolev inequalities. Finally, we extend our results beyond hyperbolic space to a class of Riemannian model manifolds $\mathbb{M}^N$ satisfying the centered isoperimetric inequality. This shows that the inequalities and methods developed in this work are robust and rely mainly on geometric and isoperimetric properties, rather than on the specific structure of hyperbolic space itself.
Read the full abstract
Number Theory
Bicomplex Algebraic Numbers
We investigate bicomplex analogues of fundamental notions from classical algebraic number theory. In particular, we show that the primitive element theorem admits a natural generalization to bicomplex extensions, giving rise to two distinct classes of extensions depending on the nature of the generating element. We further establish a key decomposition property for bicomplex extensions, which serves as a foundation for studying their rings of integers. We also observe that prime elements in the ring of integers of a number field may become semiprime in the rings of integers of suitable bicomplex extensions. Finally, we present two explicit examples of finite bicomplex extensions.
Read the full abstract
Mathematical Physics
Vertex operators, infinite wedge representations, and correlation functions of the t-Schur measure
We study the $t$-Schur measure on partitions, defined by $ \mathbb{P}(\lambda)=Z^{-1}S_\lambda(x;t)s_\lambda(y) $, where $S_\lambda(x;t)$ denotes the $t$-Schur symmetric functions and $s_\lambda(y)$ the ordinary Schur functions, and $Z$ is the normalising constant. Using vertex operator calculus, we realise $S_\lambda(x;t)$ in the charged free-fermion Fock space, yielding a $t$-deformation of the classical boson-fermion correspondence. These realisations give vertex-algebraic proofs of the $t$-Cauchy identities and $t$-Gessel identity. Building on this framework, we compute the correlation functions of the $t$-Schur measure and show that the associated point process is determinantal, with an explicit correlation kernel. The Poissonised $t$-Plancherel measure appears as a specialisation of our construction, so its correlation functions follow as a corollary. As an application, we derive the limiting distribution for the length of the longest ascent pair in a random permutation. Our results interpolate the Schur case at $t=0$, connect to the Schur-$Q$ theory at $t=-1$, and provide a probabilistic interpretation of a natural $t$-refinement of increasing subsequences via a generalised RSK correspondence.
Read the full abstract
Optimization and Control
Smoothing Meets Perturbation: Unified and Tight Analysis for Nonconvex-Concave Minimax Optimization
This paper studies smooth nonconvex-concave minimax optimization and two acceleration mechanisms for single-loop first-order methods: dual perturbation and smoothing. Although both techniques improve convergence guarantees, their relative advantages remain unclear due to the distinction between game stationarity (GS) and optimization stationarity (OS). We provide a tight characterization of their iteration complexities under both notions. We show that smoothing accelerates convergence to both GS and OS, whereas dual perturbation improves the rate only for GS and does not accelerate OS. Matching lower bounds based on hard instances establish the tightness of these rates. Motivated by this separation, we propose Perturbed Smoothed GDA, a single-loop method combining both techniques. It improves the complexity for GS over existing single-loop methods while preserving the state-of-the-art rate for OS, and further admits asymptotic convergence to 0-GS, which is not available for vanilla Smoothed GDA.
Read the full abstract
Combinatorics
Word-Representation of Melon Graphs
The notion of word-representable graphs is a generalization of comparability graphs, in which graphs are represented by words. The complexity of word-representation of a word-representable graph is captured through the representation number, whereas the corresponding concept is the permutation-representation number for comparability graphs. The graphs with the (permutation-)representation number at most two were characterized in the literature. While certain examples in the class of graphs with the (permutation-)representation number three are known, no characterization for these classes is available. In this work, we prove that the representation number of melon graphs is at most three. Further, we characterize the class of melon graphs restricted to comparability graphs and show that their permutation-representation number is also at most three. Moreover, this work characterizes the word-representable line graphs of melon graphs and establishes that their representation number is at most three.
Read the full abstract
Operator Algebras
Conjugacy co-amenability
In this note we study a natural analytic property of inclusions of groups akin to co-amenability: the property of existence of a non-compactly supported invariant state for the conjugation action of a group $G$ on the von Neumann algebra generated by the characteristic functions $\{\mathbf{1}_{gHg^{-1}}\}_{g\in G}$ viewed inside $\ell^\infty(G)$. Some interesting settings and examples of this phenomena are proved. We also comment on a consideration related to proper proximality, which motivated this property.
Read the full abstract
Representation Theory
On representations of algebras with radical square zero
We consider a new correspondence between representations of algebras with radical square zero and representations of species. We show that the stable category of representations of such algebra embeds into the representation category of the corresponding species and show how one reconstruct the Auslander-Reiten quiver of the algebra from the Auslander-Reiten quiver of the species.
Read the full abstract
Commutative Algebra
On the closed neighborhood ideal of the square of the path graph
We consider the closed neighborhood ideal of square of the path graph and study some of its algebraic and homological invariants. We compute the height, the projective dimension and the Castelnuovo-Mumford regularity. We prove that these ideals are sequentially Cohen-Macaulay and characterize when they are Cohen-Macaulay.
Read the full abstract
Optimization and Control
Stable representations of Hamilton-Jacobi-Bellman equations with infinite horizon
In this paper, for the Hamilton-Jacobi-Bellman equation with an infinite horizon and state constraints, we construct a suitably regular representation. This allows us to reduce the problem of existence and uniqueness of solutions to the Frankowska and Basco theorem from (2019). Furthermore, we demonstrate that our representations are stable. The obtained results are illustrated with examples.
Read the full abstract
Optimization and Control · Cross-listed from cs.LG
A Penalty Approach for Differentiation Through Black-Box Quadratic Programming Solvers
Differentiating through the solution of a quadratic program (QP) is a central problem in differentiable optimization. Most existing approaches differentiate through the Karush–Kuhn–Tucker (KKT) system, but their computational cost and numerical robustness can degrade at scale. To address these limitations, we propose dXPP, a penalty-based differentiation framework that decouples QP solving from differentiation. In the solving step (forward pass), dXPP is solver-agnostic and can leverage any black-box QP solver. In the differentiation step (backward pass), we map the solution to a smooth approximate penalty problem and implicitly differentiate through it, requiring only the solution of a much smaller linear system in the primal variables. This approach bypasses the difficulties inherent in explicit KKT differentiation and significantly improves computational efficiency and robustness. We evaluate dXPP on various tasks, including randomly generated QPs, large-scale sparse projection problems, and a real-world multi-period portfolio optimization task. Empirical results demonstrate that dXPP is competitive with KKT-based differentiation methods and achieves substantial speedups on large-scale problems. Our implementation is open source and available at https://github.com/mmmmmmlinghu/dXPP.
Read the full abstract
Geometric Topology
Distinguishing diffeomorphism types of relative trisections
We distinguish diffeomorphism types of relative trisections using a “capping” operation, which yields a trisection diagram of a closed 4-manifold from a relative trisection diagram. Using this operation, we give various examples of non-diffeomorphic relative trisections of the same 4-manifold with boundary. We also study how the corresponding trisected 4-manifold changes under the capping operation.
Read the full abstract
Analysis of PDEs
Oblique wave interactions in 2D steady supersonic flows of Bethe-Zel'dovich-Thompson fluids
This paper studies steady supersonic flow in a 2D semi-infinite divergent duct. We assume that the flow satisfies the slip boundary condition on the walls of the duct, and the state of the flow is given at the inlet of the divergent duct. When the fluid is a polytropic ideal gas, the problem can be reduced to some interactions of rarefaction simple waves, and the existence of a global classical solution inside the divergent duct can be established using the method of characteristics. In this paper we assume that the fluid is a nonconvex Bethe-Zel'dovich-Thompson (BZT) fluid. This type of fluid may significantly differ from polytropic ideal gases. For instance, physically admissible rarefaction shocks can occur. Depending on the oncoming flow state and the flare angles of the divergent duct, thirteen distinct types of oblique wave interactions may occur, including oblique composite waves consisting of shocks and centered simple waves. This paper systematically studies these oblique wave interactions and constructs global, piecewise smooth, supersonic solutions within the divergent duct using characteristic decomposition and hodograph transformation methods. We also obtain the detailed structures of these solutions in addition to their existence. The results and methods of this paper are also applicable to some 2D Riemann problems for gases with nonconvex equations of state.
Read the full abstract
Dynamical Systems
Convergence and combinatorics of the Reverse algorithm
We study the Reverse algorithm, a multidimensional continued fraction algorithm, which is not unimodular. We show that the Reverse algorithm is ergodic and, by proving that its second Lyapunov exponent is negative, that it is a.e. exponentially convergent. In addition to that, we attach substitutions to this algorithm and study the $S$-adic languages generated by sequences of these substitutions. The negativity of the second Lyapunov exponent implies that almost all of these languages are balanced. By a thorough study of the combinatorics of the substitutions, we are even able to obtain a concrete generic family of balanced languages that is characterized in terms of a simple condition on the underlying sequence of substitutions.
Read the full abstract
Optimization and Control
Subgradient Gliding Method for Nonsmooth Convex Optimization
We identify and analyze a fundamental limitation of the classical projected subgradient method in nonsmooth convex optimization: the inevitable failure caused by the absence of valid subgradients at boundary points. We show that, under standard step sizes for both convex and strongly convex objectives, the method can fail after a single iteration with probability arbitrarily close to one, even on simple problem instances. To overcome this limitation, we propose a novel alternative termed the subgradient gliding method, which remains well defined without boundary subgradients and avoids premature termination. Beyond resolving this foundational issue, the proposed framework encompasses the classical projected subgradient method as a special case and substantially enlarges its admissible step-size design space, providing greater flexibility for algorithmic design. We establish optimal ergodic convergence rates, $\mathcal{O}(1/\sqrt{t})$ for convex problems and $\mathcal{O}(1/t)$ for strongly convex problems, and further extend the framework to stochastic settings. Notably, our analysis does not rely on global Lipschitz continuity of the objective function, requiring only mild control on subgradient growth. Numerical experiments demonstrate that, in scenarios where the classical projected subgradient method fails completely, the proposed method converges reliably with a $100\%$ success rate and achieves orders-of-magnitude improvements in accuracy and convergence speed. These results substantially expand the scope of subgradient-based optimization methods to non-Lipschitz nonsmooth convex problems.
Read the full abstract
Probability
Some properties of G-SVIEs
In this paper, we investigated the solvability of G-SVIEs under two cases: time-varying Lipschitz coefficients and integral-Lipschitz coefficients. Using the Picard iteration method, we established the existence and uniqueness of solutions to G-SVIEs under these two conditions. Additionally, we prove the continuity of the solution with respect to parameters in parameter-dependent G-SVIEs with Lipschitz coefficients.