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)

Wednesday 29 April

Statistics Theory

Technical Note on Relating Scores of Tilted Distributions

Recent results have shown that for a linear tilt to a reference measure, the scores that would be produced under convolution with a normal variable can be expressed in terms of convolutions of the original density. Here, we extend that result to include constant negative diagonal tilts as well. The relationship follows from relating the denoisers of the two densities, which define the scores via Tweedie formula. A linear tilt results in a location shift to the score operator, while a quadratic tilt results in both a location shift and a time shift. Thus the scores of the tilted density can be understood as the scores of the original convolution process at a different location and noise level. These results are of interest to those in the score based diffusion community, and may lead to better score estimators which take advantage of these tilted score relationships.
Read the full abstract

Mathematical Physics

Dynamical delocalization in disordered 2D Chern insulators

We show the existence of energies exhibiting dynamical delocalization in discrete 2D Chern insulators perturbed by a random potential in a general setting. Our proof exploits two main features of the model: jumps in the integer value of the Chern character and continuity of averaged spectral projections in both energy and disorder parameters. This allows us to show robustness of the topological index in the presence of disorder, which, combined with existing methods to prove dynamical localization, allows us to provide detailed information on the phase diagram of the model. The novelty of our approach is that we are able to show dynamical delocalization in the disorder parameter, and not only in the energy parameter, which allows to prove Anderson metal-insulator transition even when spectral gaps close due to the strength of disorder.
Read the full abstract

Mathematical Physics

The quantum group structure of long-range integrable deformations

Quantum integrable spin chains are known to possess a large family of long-range deformations generated by the local, boost and bilocal operators. Although these deformations are well-understood on the level of the pairwise commuting charges, the underlying quantum group structures had not yet been recognised. In this paper, we provide a quantum group-theoretical description for the family of long-range deformations of arbitrary homogeneous Yang-Baxter integrable spin chains up to first order in the deformation parameter. In particular, we show that the deformations are obtained via a twist of the algebraic structure of the underlying quantum group. This twisting results in a generally non-associative algebra that has a non-trivial Drinfeld associator. The Drinfeld associator is then shown to encode the information about the long-range interaction terms for the integrable spin chain. Moreover, the deformed quantum group is shown to contain a large perturbatively associative substructure, thus ensuring the perturbative integrability of the long-range model. The deformed quantum group provides explicit expressions for the Lax operators and R-matrices of the long-range deformed models, which manifestly satisfy the RLL relation and the Yang-Baxter equation up to first order in the deformation parameter. In order to derive the results, we introduce algebra elements that we call the algebraic charge densities. As a side result, we provide a conjecture for the explicit expressions of the undeformed charge densities in terms of these algebra elements.
Read the full abstract

Mathematical Physics · Cross-listed from hep-th

A Nahm transform for rotating calorons

Rotating calorons were introduced in the context of rotating quark-gluon plasmas. They are anti-self-dual gauge fields on $\mathbb{R}^4$ that are invariant under a glide rotation. We formulate a Nahm transform which identifies rotating calorons with solutions of a delayed-differential equation. Using this transform, we prove existence of an eight-parameter family of charge 1 rotating calorons with nontrivial holonomy and rotational angle $\pi$, which we construct and visualise using a numerical implementation of the Nahm transform.
Read the full abstract

Functional Analysis

Sidon-type inequalities for $p$-adic analogues of Rademacher chaos

We prove the $\frac{2d}{d+1}$-Sidon inequality for a system of functions representing the most general extension of the Rademacher $d$-chaos to the $p$-ary case.

Optimization and Control

Efficient Graph Partitioning under Resource Constraints: A Cutting-Plane Framework for Distribution Grids

This paper presents an optimal network topology control framework using cutting-plane methods for efficient network partitioning with controllable edges. The objective is to enable real-time reconfiguration of interconnected subnetworks while ensuring radial connectivity, resource feasibility, and structured leader allocation, which provide structural foundations for distributed controllability, stability, and coordination. The problem is formulated as a mixed-integer program that integrates graph-theoretic constraints, resource flow, and network structural properties to enforce an operational hierarchy. To address the combinatorial complexity of cycle elimination and leader assignment, we propose an iterative cutting-plane framework that ensures convergence to an optimal and feasible network topology. Theoretical guarantees on optimality preservation, feasibility, and convergence are established, ensuring systematic elimination of infeasible configurations while preserving the required coordination structure. Simulations on a modified Iowa 240-bus power distribution grid demonstrate the framework's effectiveness in network reconfiguration under resource constraints. The approach achieves a median speedup of 65.4x and a best-case speedup of over 78x in a 46-switch configuration.
Read the full abstract

All authors Duong Thuy Anh Nguyen, Harsha Nagarajan, Robert Ferrando, Russell Bent, David Fobes

Mathematical Physics · Cross-listed from quant-ph

Light-Cone Structure of Propagation of Entanglement

For a wide class of bipartite systems with localized couplings, we establish existence of an effective light-cone for propagation of entanglement. This result yields a hard lower bound on the time it takes, under ideal conditions (no loss, no decoherence), to transport entanglement to a distant location (say, a node of a graph-structured quantum network), or to maintain it there.
Read the full abstract

Mathematical Physics · Cross-listed from hep-th

BPS spectra of $\operatorname{Tr}[\Psi^p]$ matrix models for odd $p$

We study the BPS cohomology and Hamiltonian of a model built from an $N\times N$ matrix $\Psi$ of complex fermions, with supercharge $Q_p={\rm Tr}(\Psi^p)$ for odd $p\ge3$. Numerical calculations give the BPS multiplicity at every fermion number for $(p,N)=(5,3),(5,4),(5,5),(7,4)$, and through fermion number six at $(7,5)$. In $Z_{BPS}^{(p,N)}(x)=\sum_Rh_Rx^R$, $x$ records fermion number and $h_R$ counts BPS states in sector $R$. In each of the four cases computed at every fermion number, factoring out the lowest power of $x$ leaves a polynomial divisible by a positive power of $p$ and by $(1+x)^N$. The quotient by these factors has nonnegative integer coefficients. Exact decoupling of ${\rm Tr}\Psi$ and the Euler characteristic prove divisibility by $(1+x)^2$. Pairing sectors of complementary fermion number gives a third factor when $N$ is odd. For odd $p\le2N-1$, the only range in which $Q_p$ can be nonzero, the same pairing proves the corresponding rank symmetry for $Q_p$. The additional factors required to reach $(1+x)^N$ remain conjectural in general. Evaluating at $x=1$ gives the total BPS count $Z_{BPS}^{(p,N)}(1)$. For each fixed odd $p$, the Witten indices imply that, for any sequence of matrix sizes $N_j\to\infty$ along which $N_j^{-2}\log Z_{BPS}^{(p,N_j)}(1)$ converges, its limit lies in $[\log(2\cos\frac{\pi}{2p}),\log 2]$. After dividing the Hamiltonian by $p^2$, normal ordering gives an alternating sum of operators $K_k$, with $K_k$ containing $k$ creation and $k$ annihilation operators. We obtain exact formulas for $K_0,K_1,K_2$ when $p=5$ and $N\ge3$. Only $K_3,K_4$ can contain spectral information beyond the number of traceless fermions and the quadratic Casimir. At $N=3$, the Hodge star maps the invariant cubic form to the quintic form up to scale, so all five terms commute. Calculations with integer matrices give nonzero commutators in the reported $N=4$ sectors.
Read the full abstract

Commutative Algebra

Reverse Tableaux and the Surjectivity of the Component Map in Type $A$

Let $G = \mathrm{SL}(n,\mathbb{C})$, let $B$ be a fixed Borel subgroup, and let $P \supset B$ be a parabolic subgroup determined by a composition $(c_1,\dots,c_k)$ of $n$. Write $P'$ for the derived group of $P$ and $\mathfrak{m}$ for the Lie algebra of the nilradical of $P$. By Richardson's theorem the algebra of semi-invariants $\mathscr{I} := \mathbb{C}[\mathfrak{m}]^{P'}$ is polynomial; in type $A$ its generators may be taken to be the Benlolo–Sanderson (BS) invariants. The nilfibre is the common zero locus $\mathscr{N} := V(\mathscr{I}_{+}) \subset \mathfrak{m}$. A set of component tableaux, each encoding combinatorial data summarised in a multi-set called the Red Set, was constructed in earlier work by Y. Fittouhi and A. Joseph in The reverse tableau: a gateway to the surjectivity of the component map. The resulting component map $\phi : \{\text{component tableaux}\} \to \Irr(\mathscr{N})$ was shown to be injective. In the present article, we develop the Factorization Principle for Benlolo–Sanderson invariants in order to give a rigorous proof of the surjectivity of the component map $\phi$. While the combinatorial framework of reverse tableaux was introduced in a work by Y. Fittouhi and A. Joseph cited above, the surjectivity of $\phi$ remained conjectural: the linearization method used there did not exclude the possible loss or merging of irreducible components. The present paper resolves this geometric difficulty by showing that the relevant invariants factorize into products indexed by pseudo-neighbouring column pairs, thereby ensuring that every component is reached in a controlled and accountable way.
Read the full abstract

Numerical Analysis

Embeddings of Reproducing Kernel Hilbert Spaces with General Weights

We study embeddings between reproducing kernel Hilbert spaces $H(K)$ of functions of $d \in \mathbb{N} \cup \{\infty\}$ variables. The kernels $K$ are superpositions of weighted finite tensor products of a fixed univariate kernel. The basic idea for the embeddings is to compensate a change of the univariate kernel by a suitable transformation of the weights. For the proofs we employ ($d \in \mathbb{N}$) and develop ($d = \infty$) a discrete calculus on the cone of all weights, where completely monotone weights play a particular role. We sketch how to apply the embedding results to computational problems, as, e.g., numerical integration or function recovery.
Read the full abstract

Analysis of PDEs

On the unimportance of distant players in sparse stochastic differential network games

We study stochastic differential games with $N$ players, where interactions are determined by sequences of graphs in which the number of neighbours of each node remains bounded as $N$ grows, such as chain graphs or lattices. Our main goal is to quantify the phenomenon of the "unimportance of distant players" in such a large population, sparse regime: we show that, in order to determine the optimal trajectory in open-loop strategies of a given player with an arbitrarily small error, it suffices to consider a reduced game involving only the players at a certain distance in the graph, assigning arbitrary trajectories to the farther ones. Our main result provides an explicit non-asymptotic estimate in terms of the graph distance, valid independently of the time horizon $T$, under suitable convexity and monotonicity assumptions on the costs. Similar results are obtained for games in distributed strategies.
Read the full abstract

Optimization and Control

A Scale-Shape Dual Newton Method for Entropic Least Squares

We give a damped inexact Newton method for entropy-regularized least-squares on the nonnegative orthant that converges globally at a linear rate with $O(\log\epsilon^{-1})$ iteration complexity, locally at a superlinear-to-quadratic rate, and is immune to the finite-precision overflow that limits classical dual solvers. A scale-shape decomposition of the primal – separating its scale from its direction – produces a dual with a nonsingular Jacobian. Objectives and Jacobians are evaluated through stable log-sum-exp and softmax primitives. Lambert W bounds on the scale uniformly control the Jacobian's spectrum, from which both rates follow. The solution map is jointly Lipschitz in the data, regularization parameter, and reference measure, and extends continuously to the vanishing-regularization limit. Experiments on a problem from analytic continuation of quantum Monte Carlo data confirm the predicted overflow resilience and convergence behavior.
Read the full abstract

All authors Nicholas Barnfield, James V. Burke, Michael P. Friedlander, Tim Hoheisel

Algebraic Geometry

Characterization of non-special divisors of small degree on Kummer extensions and LCP codes

A recent construction of linear complementary pairs (LCPs) of algebraic geometry codes is intimately linked to the identification of non-special divisors of small degree within a function field over a finite field. Let $\mathbb{F}_q$ be the finite field of cardinality $q$. In this work, we consider a function field $F/\mathbb{F}_q$ of genus $g$ defined by a Kummer extension of type $y^m = f(x)$, where $f(x)$ is a polynomial in $\mathbb{F}_q[x]$. Based on the theory of generalized Weierstrass semigroups at several places, we provide an arithmetic criterion to identify all non-special divisors of degree $g-1$ and $g$ whose support is contained in a subset of the totally ramified places of the extension $F/\mathbb{F}_q(x)$. Furthermore, we explicitly determine all non-special divisors of degree $g-1$ in certain cases. Finally, we apply these results to provide explicit new families of LCPs algebraic geometry codes.
Read the full abstract

Combinatorics

Hamilton decompositions of the directed 5-torus for odd modulus

We prove that the directed five-dimensional torus $D_5(m) = \operatorname{Cay}((\mathbb{Z}_m)^5, \{e_0, e_1, e_2, e_3, e_4\})$ has a Hamilton decomposition for every odd integer $m \geq 3$. This is the first higher-dimensional case in which the return-map method requires a genuine zero-set selector rather than an odometer-type correction. The construction assigns the five outgoing generators by a cyclic layer schedule with one non-constant layer determined by a zero-set Latin table; an explicit finite exact-cover certificate proves that this layer is a matching. By cyclic symmetry, Hamiltonicity of all color classes reduces to a single normalized return map. For $m \geq 5$, an explicit first-return calculation on the section $p = 2$ gives one induced cycle whose excursion lengths sum to $m^4$. The remaining modulus $m = 3$ is settled by a printed finite cycle certificate. A companion Lean 4 formalization provides an independent machine verification of the Cayley statement and the finite certificates; source, audit scripts, and ancillary search code are available at https://github.com/aria1th/Torus-Hamilton-Decomposition-Program.
Read the full abstract

Optimization and Control

Explainable Artificial Intelligence for Financial Integral Equations: A Fixed-Point Neural Operator Approach

The explainable artificial intelligence is used to analyze the stochastic Fredholm integral equations (SFIEs) and stochastic deep neural networks (SDNNs). The neural operator-based stochastic fixed point framework is used to develop SDNNs. The solution of an SFIE is obtained through successive applications of an integral operator, and this iterative structure naturally resembles the layered architecture of a neural network. The associated nonlinear versions of SFIE and SDNN are discussed. The SFIE and SDNN are used to solve the Black-Scholes equation, contagion dynamics of financial networks, and the Merten jump diffusion equation. It is observed that the results obtained through SFIE and SDNN for all the applications agree well.
Read the full abstract

Probability

Applied Random Matrix Theory

Random matrices now play a role in many parts of computational mathematics. To advance these applications, it is desirable to have tools that are flexible, easy to use, and powerful. Over the last 25 years, researchers have developed a remarkable family of results, called matrix concentration inequalities, that meet the criteria. This paper offers an invitation to the field of matrix concentration and its multifarious applications.
Read the full abstract

Dynamical Systems · Cross-listed from q-bio.PE

Host diversity and adaptive vector feeding preferences shape disease burden in vector-borne diseases

Vector-borne diseases often involve multiple host species that differ in their ability to sustain transmission. At the same time, vector feeding preferences can change in response to host availability and disease-control interventions, potentially altering disease dynamics in unexpected ways. We develop a two-host vector-borne disease model that links host diversity, adaptive vector feeding preferences, and disease transmission. We demonstrate that the effect of host diversity on disease transmission is mediated by vector feeding behavior and cannot be inferred from host abundance alone. In particular, we identify a critical threshold, $R_{0c}$, that determines whether shifts in vector preference amplify or suppress disease burden in a focal host. This threshold marks a qualitative transition in system behavior and provides a basis for predicting epidemiological responses to changes in host composition and vector behavior. Using adaptive dynamics, we further show that vector populations may evolve toward either specialist or opportunistic feeding strategies depending on host encounter rates and trade-off strength. Finally, we demonstrate that host-targeted interventions can induce adaptive changes in vector feeding behavior that reduce prevalence in the protected host while potentially increasing overall infection burden. Our results highlight how evolutionary responses of vector population can generate unexpected epidemiological outcomes and should be considered when designing disease-control strategies.
Read the full abstract

Number Theory

An $O_K$-basis for the image of a Lubin-Tate logarithm on $\pi$-regular extensions of $K$

Let $K$ be a finite $p$-adic field with uniformiser $\pi$. In this paper we study the image of the logarithm attached to a Lubin-Tate series $[\pi](X)$ on the maximal ideal of so-called $\pi$-regular extensions of $K$; for such an extension $L|K$ we compute a basis for the additive group $\log_{[\pi]}(\mathcal{F}(\mathfrak{m}_L))$ as an $O_K$-module, where $\mathcal{F}(\mathfrak{m}_L)$ denotes the maximal ideal $\mathfrak{m}_L$ equipped with the $O_K$-module structure coming from the formal group associated to $[\pi](X)$, and determine the minimal valuation of the elements in $\log_{[\pi]}(\mathcal{F}(\mathfrak{m}_L))$. In the final section of this paper we discuss how some of these results extend to arbitrary finite extensions of $K$ and conclude by determining a basis of the $O_K$-module $\log_{[\pi]}(\mathcal{F}(\mathfrak{m}_{K_{\pi^n}}))$, where $K_{\pi^n}$ is the Lubin-Tate extension of level $n\geq 1$.
Read the full abstract

Functional Analysis

Weakly, sufficiently or strongly localized operators on the Fock space in \mathh C^n

We study properties of the following four classes of operators on the Fock space in $\mathbb C^n:$ 1) weakly localized operators; 2) sufficiently localized operators in the sense of Xia and Zheng; 3) sufficiently localized operators; 4) strongly localized operators. In this respect, we examine composition operators, Toeplitz operators with a measure symbol whose total variation measure is a Fock-Carleson measure, and singular operators of convolution type introduced by Zhu, among others. We also provide a bounded operator which is not weakly localized and does not even belong to the Toeplitz algebra. Class 1) contains class 2), class 2) contains class 3), which clearly contains class 4). We prove that the first two inclusions are strict. Our proofs are in terms of singular operators of convolution type introduced by Zhu. The third inclusion was already known to be strict, as Wang, Cao and Zhu exhibited examples of composition operators which are sufficiently localized, but are not strongly localized. %As our main result, we show the existence of a singular operator of convolution type which is weakly localized, but is not sufficiently localized in the sense of Xia and Zheng. %The underlying question is whether the first two classes of operators coincide or not.
Read the full abstract

All authors David Békollè, Solange B. Difo, Hugues O. Défo, Edgar L. Tchoundja

Combinatorics

Stretched Schubert coefficients are eventually quasi-polynomial

For a permutation $u\in S_n$, let $N\ast u\in S_{Nn}$ be the permutation with scaled Lehmer code. For given $u,v,w\in S_n$ and integer $N$, the stretched Schubert coefficients are defined as $f_{u,v,w}(N):=c_{N*u,N*v}^{N*w}$. Our main result is that the function $f_{u,v,w}(N)$ is eventually quasi-polynomial. This proves Kirillov's conjecture (2004), that the generating function for the sequence $\{f_{u,v,w}(N)\}$ is rational. For the proof, we use combinatorics of pipe dreams to show that Schubert coefficients are given as an alternating sum of the numbers of integer points in certain polytopes. These polytopes behave nicely under stretching, and we use Ehrhart theory to obtain the result. As a consequence of the proof, we also present new counterexamples to the saturation conjecture for Schubert coefficients, and give computational applications.
Read the full abstract

Information Theory

Low-Complexity Run-Length-Limited ISI-Mitigation (RLIM) Codes for Molecular Communication

Molecular communication suffers from severe inter-symbol interference, which makes constrained coding essential for reliable transmission. Run-length-limited ISI-mitigation codes are attractive because they select low-weight constrained codebooks, reducing ISI while allowing more molecules to be assigned to each transmitted \(1\)-symbol under the usual molecular-communication normalization. Previous results showed strong bit-error-rate performance for these codes, but their original realization required full codebook generation and storage. This exponential storage growth is unsuitable for resource-constrained molecular communication channels and also limits the exploration of larger information dimensions. This is particularly important for nano-scale molecular communication, where transmitter and receiver nodes are expected to operate under severe memory and computational constraints. This paper removes that realization bottleneck by replacing full codebook storage with an enumerative realization based on Cover's ranking framework, constant-weight run-length-limited counting, and cumulative weight-layer offsets. The resulting encoder and decoder preserve the selected RLIM codebooks and the original projection-based decoding behavior while storing only polynomial-size counting tables. Storage and runtime measurements confirm the resulting exponential-to-polynomial reduction, while diffusion-based molecular-communication simulations show that the newly accessible larger information dimensions can improve the best mean bit-error-rate performance attainable by individual RLIM orders in several tested channel regimes.
Read the full abstract

Combinatorics

Polynomial and spectra factorization of graphs obtained by iteration the operad of generalized graph composition

The generalized composition graph is used by Cardoso and some researchers for factorization of the adjacency spectrum and Laplacian of a simple graph. Because the generalized composition graph is an example of a set-theoretic linear operad, this operation can be iterated at more than one level, where the complex language of partition refinement in the iteration is represented in terms of Schr"oder trees. This allows us to generalize the factorization of the adjacency spectrum and Laplacian of a simple graph presented by Cardoso in terms of Schr"oder trees and colorings over the edges of a graph. Cardoso's technique has been generalized by other authors for the universal adjacency matrix of a graph. This work also presents generalized factorizations in terms of Schr"oder trees and colorings on the edges of a graph for the universal adjacency spectrum, the characteristic polynomial of the universal adjacency matrix, and the generalized characteristic polynomial of a graph.
Read the full abstract

Representation Theory · Cross-listed from hep-th

Oscillators from non-semisimple walled Brauer algebras

The walled Brauer algebras $B_N(m,n)$ govern Schur–Weyl duality for unitary groups $U(N)$ acting on mixed tensor spaces $V_N^{\otimes m}\otimes \overline{V}_N^{\otimes n}$ and play an important role in applications ranging from AdS/CFT to quantum information theory. In the stable regime $N\ge m+n$ the algebra is semisimple and its representation theory is well understood. For $N<m+n$, however, $B_N(m,n)$ becomes non-semisimple. The representation of the algebra on tensor space has a non-trivial kernel and the corresponding quotient algebra is semisimple, with representation dimensions differing from those in the stable regime. We introduce restricted Bratteli diagrams, obtained by modifying the standard Bratteli diagrams for $B_N(m,n)$. This construction provides a systematic way to use representation-theoretic data from the stable regime to compute the dimension modifications arising in the non-semisimple regime. In the regime $N=m+n-l$, with $l$ small compared to $m,n$, we show that the restricted diagrams exhibit a stability property and enable an efficient counting of the paths responsible for these dimension corrections. Remarkably, the resulting generating functions are governed by the partition function of an infinite tower of simple harmonic oscillators. We briefly discuss implications for the construction of orthogonal bases of matrix invariants in gauge theory and related applications in quantum information theory.
Read the full abstract

Differential Geometry

Symplectic structure on the character varieties of Sasakian threefolds

Take a compact Sasakian threefold $M$ and consider the associated irreducible $\text{SL}(r,{\mathbb C})$-character variety ${\mathcal R} := \text{Hom}(\pi_1(M, x_0), \text{SL}(r, {\mathbb C}))^{ir}/ \text{SL}(r, {\mathbb C})$ of $M$, where $\text{Hom}(\pi_1(M, x_0), \text{SL}(r, {\mathbb C}))^{ir}$ is the space of irreducible homomorphisms. We first construct a natural algebraic $2$-form on $\mathcal R$. Then it is shown that this $2$–form is closed. Finally we show that the restriction of this $2$–form to $\text{Hom}(\pi_1(M, x_0), \text{SU}(r))^{ir}$ is symplectic.
Read the full abstract

Symplectic Geometry

Holomorphic disks and tropical Lagrangians

We develop a calculus for counting pseudoholomorphic disks with boundary in tropical Lagrangians contained in almost toric manifolds, using our previous work with Venugopalan. The results are mostly in dimension four under monotonicity assumptions although in principle the same technique works in any dimension and without monotonicity. The calculus is given as a sum over tropical graphs that interact with the tropical graph of the Lagrangian, generalizing results of Mikhalkin and Nishinou-Siebert for holomorphic spheres in toric varieties, and our previous result with Venugopalan which dealt with disks bounding almost toric moment fibers. The main contribution of this paper is the calculation of several multiplicities of vertices corresponding to disks, such as the holomorphic pant (half of the holomorphic pair of pants) and various univalent vertices occuring at trivalent vertices of the graph of the Lagrangian; a key tool is a Lagrangian isotopy from the Lagrangian pair of pants in the del Pezzo of degree seven to the inverse image of a diagonal, which is a special case of a results of Hind and Evans. We show that every integer eigenvalue of non-maximal modulus for quantum multiplication by the first Chern class is realized by such a sphere.
Read the full abstract

Optimization and Control

Nonsmooth Riemannian optimization with inexact manifold primitives via bundle methods

Optimization on Hadamard manifolds – the natural Riemannian setting for globally geodesically convex problems – relies on exponential maps to retract tangent vectors and parallel transport to connect tangent spaces across the manifold. These primitives are often computationally expensive, leading software packages to rely on approximations: first-order retractions and vector transports. However, existing results for optimization on Hadamard manifolds either require exact primitives or lack non-asymptotic rates. We bridge this gap by introducing a proximal bundle method for nonsmooth geodesically convex optimization and establishing the first oracle-complexity bounds that rely only on subgradients and inexact primitives. We obtain sublinear rates for general objectives and optimal linear convergence under sharp function growth.
Read the full abstract

Information Theory

On Coded Caching Systems with Decentralized Linear Coding Placement

Coded caching is a technique that leverages locally cached contents at the end users to reduce the network's peak-time communication load. Coded caching has been shown to achieve significant performance gains with a centralized placement orchestrated by the server and is thus considered a promising technique to boost performance in future networks by effectively trading off bandwidth for storage. To tackle issues caused by the synchronized placement, previous works focused on decentralized placement and found the exact worst-case load with uncoded placement. In this paper, we focus on a decentralized coded caching system with random linear coding placement, and investigate the fundamental limits of a linear coding placement where each user independently and uniformly caches random linear coding symbols of a single file. We propose achievable and converse bounds on the worst-case load, which are shown to meet under certain conditions.
Read the full abstract

Mathematical Physics · Cross-listed from hep-th

Perturbative Coulomb branches on $\mathbb{R}^3\times S^1$: the global D-term potential

We develop a Wilsonian description of perturbative D-terms in 4d $\mathcal N=1$ gauge theories on $\mathbb R^3\times S^1$ with chiral Kaluza–Klein spectra. A closed-form, one-loop-exact coefficient of the auxiliary D fields, including 4d Fayet–Iliopoulos parameters and globally valid throughout the holonomy space, determines the leading holonomy potential away from massless thresholds. We establish uniform weak-coupling control there and match across thresholds by retaining the light fields. Gauge-anomaly cancellation renders the Coulomb-slice vacuum equations affine within each chamber bounded by the thresholds, enabling exact linear analysis of their zero sets and numerical reconstruction through DBSCAN clustering of approximate zeros, LPCA denoising, and RANSAC hyperplane detection. Complementing the rank-one applications of the accompanying Letter, we study flavor-twist dependence in the $\mathrm{SU}(2)$ $\mathcal N=4$ theory, and illustrate the numerical reconstruction in higher-rank R-twisted Argyres–Douglas reductions of interest in the SCFT/VOA correspondence. We also relate the same D-term data to leading stationary-phase equations of supersymmetric partition functions and a decompactification limit of the torus twisted superpotential.
Read the full abstract

Functional Analysis

Operator-Valued Positivstellensätze on Matrix Convex Sets and Free Products of Finite Abelian Groups

We prove a Positivstellensatz for operator-valued noncommutative polynomials that are positive on matrix convex sets. Specifically, let $p$ be an operator-valued polynomial in $B(H)\otimes C<x>$ of degree at most $2d+1$, where $H$ is separable and infinite-dimensional. Let $L(x)=I+\sum_{j=1}^{g} A_j x_j$ be a monic linear operator pencil, and let $D_L=\{X: L(X) \geq 0\}$ be the associated matrix convex set. We show that $p$ is positive on $D_L$ if and only if $p=r^*r+q^*\pi(L)q$, where $q$ and $r$ have degree at most $d$, and $\pi$ is a unital completely positive map on the operator system generated by the coefficients of $L$. The proof combines a Hahn–Banach separation argument with a tailored GNS construction. The main challenge is that the separation occurs in the product ultraweak topology, so boundedness of the resulting GNS operators is not automatic. We first handle bounded matrix convex sets, using closedness of the cone of weighted squares in the product ultraweak topology as the key technical input, and then pass to the general unbounded case by an approximation argument. Finally, we apply this convex Positivstellensatz to prove an operator-valued noncommutative Fejer–Riesz theorem on free products of finite abelian groups. The key additional ingredients are the universal $*$-algebra povm(n) associated with POVMs, a perfect Positivstellensatz for povm(n), and Boca's theorem on free products of completely positive maps. As a consequence, every positive operator-valued trigonometric polynomial on a free product of finite abelian groups admits a sum-of-squares factorization with explicit complexity bounds.
Read the full abstract

Mathematical Physics · Cross-listed from nlin.SI

Scalene Yang-Baxter maps and Lax triples

We study a generalisation of the set-theoretic Yang-Baxter equation and investigate the connection between its solutions and matrix refactorisation problems. We refer to such solutions as scalene Yang-Baxter maps. Moreover, we construct scalene Yang-Baxter maps associated with integrable equations of KdV and NLS type.
Read the full abstract