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)

Thursday 30 April

Probability

Time-dependent Robin heat equation via Markovian switching

This paper investigates the heat equation on a bounded domain with a Robin boundary condition, where the reactivity parameter (or killing rate) is modeled as a continuous-time Markov chain. We analyze the system under two stochastic frameworks using a functional analytic approach. First, we examine the annealed case, which accounts for the joint stochasticity of the diffusion and the switching mechanism. We describe the solution via a strongly continuous contraction semigroup on a product space. We identify its infinitesimal generator, which incorporates the state-dependent Robin conditions into its domain, and provide a corresponding Feynman-Kac formula. Second, we study the quenched setting for fixed realizations of the switching paths. We characterize the solution through a non-autonomous evolution family (propagator) and derive a Feynman-Kac-type representation involving the boundary local time of a reflected Brownian motion. We prove an averaging principle in the fast-switching limit, showing that the system converges to a deterministic Robin problem. These results are applied to a biophysical model of stochastically gated receptors on cell membranes.
Read the full abstract

Algebraic Geometry

Valuative independence for Calabi–Yau varieties

We construct valuatively independent bases for the space of sections of an ample line bundle on a log Calabi–Yau pair over a discretely valued field and the space of regular functions on an affine CY pair with maximal boundary. While the bases are not in general unique, they induce canonical functions on the respective skeletons and are expected to agree with tropicalizations of theta functions when they exist. The proof uses techniques from the study of higher rank degenerations in K-stability.
Read the full abstract

Statistics Theory

Decoupled Descent: Exact Test Error Tracking Via Approximate Message Passing

In modern parametric model training, full-batch gradient descent (and its variants) suffers due to progressively stronger biasing towards the exact realization of training data; this drives the systematic “generalization gap”, where the train error becomes an unreliable proxy for test error. Existing approaches either argue this gap is benign through complex analysis or sacrifice data to a validation set. In contrast, we introduce decoupled descent (DD), a novel theory-based training algorithm that satisfies a train-test identity – enforcing the train error to asymptotically track the test error for stylized Gaussian mixture models. Within this specific regime, leveraging approximate message passing theory, DD iteratively cancels the biases due to data reuse, rigorously demonstrating the feasibility of zero-cost validation and $100\%$ data utilization. Moreover, DD is governed by a low-dimensional state evolution recursion, rendering the dynamics of the algorithm transparent and tractable. We validate DD on XOR classification, yielding superior performance compared to GD; additionally, we implement noisy MNIST and non-linear probing of CIFAR-10, demonstrating that even when our stylized assumptions are relaxed, DD narrows the generalization gap compared to GD.
Read the full abstract

Analysis of PDEs

Nonlocalized damping estimates for hyperbolic systems of balance laws in one space dimension

In this paper, we present a new approach to obtain so-called damping estimates for self-similar solutions to general hyperbolic relaxation systems. Such damping estimates are an important part of the stability theory of relaxation profiles with and without subshock where they enable the closure of nonlinear stability arguments. We extend the damping estimates obtained in Mascia and Zumbrun (2005) and Yang and Zumbrun (2020) from the $L^2$-case to the $L^\infty$-case and, at the same time, generalize the $L^2$-estimates to the general non-symmetric setting. Compared to previous deductions of damping estimates, the method chosen here is considerably simpler and avoids delicate Kawashima-type estimates (Mascia and Zumbrun (2005) and Yang and Zumbrun (2020)) and the use of pseudo-differential calculus (Zumbrun 2026). The novel nonlocalized damping estimates open the door to a general stability theory of shock profiles of hyperbolic relaxation systems under nonlocalized perturbations in one space dimension.
Read the full abstract

Algebraic Topology

On the homotopy types of $4$-dimensional toric orbifolds

The cohomological rigidity problem for toric orbifolds asks when an integral cohomology isomorphism implies a homotopy equivalence. In this paper we reformulate the cohomological rigidity problem in the context of $4$-dimensional toric orbifolds by introducing what we call proper isomorphisms, a variant of a concept studied by J.H.C. Whitehead. We prove that each proper isomorphism class of $4$-dimensional toric orbifolds contains at most two distinct homotopy types, and that the two classifications agree in certain special circumstances.
Read the full abstract

Combinatorics

Irreducible Ferrers diagrams in the Etzion-Silberstein conjecture

The Etzion-Silberstein conjecture asserts that, for any finite field $\mathbb F$, Ferrers diagram $\mathcal D$, and integer $d$, there exists a linear matrix code supported on $\mathcal D$ with minimum rank distance $d$ that attains a natural upper bound on its dimension. Codes achieving this bound are called maximum Ferrers diagram (MFD) codes. While the conjecture has been established for several classes of diagrams (including rectangular, monotone, and MDS-constructible cases), it remains open in general. In this paper, we study the reducibility of Ferrers diagrams. For a fixed distance $d$, a diagram $\mathcal D$ is said to reduce to $\mathcal D'$ if an MFD code for $(\mathcal D,d)$ can be obtained from one for $(\mathcal D',d)$ via shortening or inclusion. Diagrams that are not reducible are called irreducible. We show that the conjecture holds for all diagrams if and only if it holds for irreducible ones, thereby reducing the problem to this fundamental class. Our main result provides a complete characterization of irreducible diagrams: for each $d$, they correspond exactly to the integer points of a polytope $\mathfrak{P}_d \subset \mathbb{R}^{2d-3}$. We prove that these polytopes are integral, enabling the use of Ehrhart-theoretic tools to study their structure. Finally, we formulate a new conjecture on puncturing and inclusion of maximum rank distance codes, and show that it arises as a special case of the Etzion-Silberstein conjecture.
Read the full abstract

Combinatorics

A Ring structure on the Class of Combinatorial Games

J. Conway defined useful operations on the Class of combinatorial games and also introduced a notion of equivalence between games. Conway showed that, under his equivalence, games form a Group. However, Conway product is not well defined on equivalence classes of arbitrary games (though it is well defined for surreals). We consider an equivalence relation finer than Conway's and show that under such a relation combinatorial games actually form a Ring. We hint to other possible relations on the Class of combinatorial games.
Read the full abstract

Functional Analysis

On the free Banach lattice generated by a lattice

We study structural properties of the free Banach lattice $FBL\langle L\rangle$ generated by a distributive lattice $L$. We characterize when $FBL\langle L\rangle$ has a strong unit, compute its density character, analyze the density character of order intervals and study when is $FVL\langle L\rangle$ order dense in $FBL\langle L\rangle$. We also study projection bands, quasi-interior points, and Banach lattice homomorphisms induced by lattice homomorphisms. Finally, we show that $FBL\langle L\rangle$ is lattice isometric to $FBL\langle L^{\mathrm{op}}\rangle$, where $L^{\mathrm{op}}$ denotes the opposite lattice.
Read the full abstract

Functional Analysis

Uncentred maximal operators with respect to half balls on Damek–Ricci spaces

In this paper we study a variant of the uncentred Hardy–Littlewood maximal operator on Damek–Ricci spaces in which balls are replaced by suitable half balls. Perhaps surprisingly, such modified maximal operator has better boundedness properties than the classical one. In particular, it is bounded on $L^p$ for every $p$ in $(1,\infty]$ (whereas the analogue operator on balls is bounded on $L^p$ only for $p>2$), and satisfies a limiting distributional inequality if $f$ is in $L\log (2+L)$. This endpoint estimate is optimal in the sense that it does not hold if $L\log ({2}+L)$ is replaced by a larger (in a suitable sense) Orlicz space.
Read the full abstract

All authors Nikolaos Chalmoukis, Stefano Meda, Effie Papageorgiou, Federico Santagati

Mathematical Physics

Hypergeometric Functions of Nilpotent Operators: Functional Collapse and Structural Depth at Exceptional Points

We study hypergeometric functions of nilpotent operators in finite-dimensional settings, motivated by the algebraic structure of exceptional points in non-Hermitian quantum mechanics. Our starting point is the following exact result: if N is a nilpotent operator of index m+1 in an associative algebra over C, then every generalized hypergeometric function pFq evaluated at N reduces to a finite polynomial in N of degree at most m, without any analytic convergence requirement. This "functional collapse" is distinct from the classical parameter-termination mechanism and arises purely from the nilpotent structure of the argument. The main result is a "nilpotent depth criterion" (Theorem 2): if the first non-constant coefficient of a formal series F appears in degree r >= 1, then the nilpotent part F(N) - F(0)I has nilpotency index bounded above by ceil((m+1)/r). We apply this criterion to Hamiltonians at exceptional points, where H = lambda I + N with N^{m+1} = 0. Theorem 3 establishes that a function F analytic at lambda reduces the Jordan depth of the exceptional point from m+1 to at most ceil((m+1)/r), where r is the contact order of F at lambda. As consequences: the time evolution operator e^{tH} preserves the full Jordan depth for all t != 0; a function with a zero of order m+1 at lambda annihilates the entire Jordan structure; and the order of the pole of the modified resolvent is reduced from m+1 to at most m+1-r. Results are illustrated with explicit 3x3 Jordan block computations for 1F1, 2F1, and the time evolution operator, confirming sharpness of the bounds.
Read the full abstract

Dynamical Systems

Two remarks on transcendental shift-like maps on $\mathbb{C}^N$

In \cite{Bedford}, the dynamics of a particular polynomial diffeomorphism of $\mathbb{C}^N$, called a polynomial shift-like map of type $\nu$, has been studied as a higher dimensional analog of Hénon maps. In this note, we prove that the Julia set of their transcendental counterpart is non-empty. In addition, an example of a transcendental shift-like map with an escaping wandering domain has been provided which, in fact, showcases a contrast with the dynamics of a polynomial shift map.
Read the full abstract

Information Theory · Cross-listed from quant-ph

High-Girth Regular Quantum LDPC Codes from Square-Base Hypergraph Products via CPM Lifts

We study square-base Calderbank–Shor–Steane (CSS) hypergraph-product codes as a finite-length class for regular high-girth quantum low-density parity-check (LDPC) design. For base matrices of small column weight, we give checkable conditions for regularity, rank deficiency, and short-cycle exclusion, and we present explicit column-weight-three and column-weight-four examples with Tanner girth 6 and 8. We also analyze circulant permutation matrix (CPM) lifts of this class. Using the standard voltage-sum criterion, we identify orthogonality-forced Tanner 8-cycles and show that CPM lifting cannot raise the Tanner girth beyond 8 when these cycles are present. As a representative finite-length instance, a randomized CPM lift of the girth-8 base construction gives a $[[28800,62,18\le d\le192]]$ girth-8 $(3,6)$-regular CSS-LDPC code. Complete enumeration excludes nontrivial logical supports of weight at most $16$ on both sides, giving $d_X,d_Z\ge18$; explicit weight-$192$ representatives give $d_X,d_Z\le192$. Under degeneracy-aware belief-propagation decoding with optional ordered-statistics-decoding-lite post-processing, this code produced zero decoding failures in $2.993\times 10^8$ independent trials at depolarizing probability $p=0.1402$; the Wilson 95% upper confidence bound is $1.28\times 10^{-8}$.
Read the full abstract

Logic

VC-Density in Divisible Oriented Abelian Groups and Their Pairs

We show that the VC-density in certain theories of oriented abelian groups is at most the size of parameter variables, which yields dp-minimality. We further prove that the VC-density of formulas in pairs of such models is bounded by twice the size of parameter variables. This uniform upper bound is shown to be sharp, and as a consequence, we show that such pairs have dp-rank 2.
Read the full abstract

Statistics Theory

A High Dimensional Wild Bootstrap Max-Test for Detecting the Presence of Significant Predictors

We construct a block bootstrap max-test for detecting the presence of significant predictors in a high dimensional setting, allowing for weakly dependent and heterogeneous (possibly non-stationary) data. The number of covariates to be screened may be large $p$ $>>$ $n$, and growing at an exponential rate, provided $\ln (p)$ $=$ $o(n^{a})$ for some $a$ $>$ $0$ that depends on memory decay and the growth of higher moments. We study the problem of correlation screening in a high dimensional marginal regression setting, assuming so-called physical dependence in a time series setting. We entirely sidestep covariance matrix estimation and adaptive re-sampling by working with a max-statistic over the many computed parameters. Thus we do not need endogenous selection of the most relevant predictor index yielding non-uniform asymptotics, nor do we need a post-estimation Bonferroni correction. The sequence of non-standard approximating distributions arising from the maximum of an increasing number of estimators is easily approximated by a multiplier (wild) block bootstrap. The max-test controls for size well, performs well against various deviations from the null, including very slight deviations with a weak or sparse signal. A numerical experiment is performed and an empirical example with the VIX volatility index is provided.
Read the full abstract

Combinatorics · Cross-listed from cs.DM

Separating Feasibility and Movement in Solution Discovery: The Case of Path Discovery

We study solution discovery, where the goal is to obtain a feasible solution to a problem from an initial configuration by a bounded sequence of local moves. In many applications, however, the graph that defines which vertex sets are feasible is not the same as the graph that governs how tokens, agents, or resources may move. Existing models such as token sliding and token jumping typically do not distinguish the problem graph and the movement graph. Motivated by this mismatch, we introduce a directed weighted two-graph model that cleanly separates feasibility from movement. A problem graph specifies the desired combinatorial objects, while a movement graph specifies admissible relocations and their costs. This yields a flexible framework that captures asymmetry, heterogeneous movement constraints, and weighted transitions, while subsuming classical discovery models as special cases. We investigate this model through Path Discovery and Shortest Path Discovery, where the task is to realize a vertex set containing an $s$-$t$-path or a shortest $s$-$t$-path in the problem graph. These problems are particularly natural in applications, since directed and weighted shortest paths are among the most fundamental algorithmic primitives. At the same time, previous work has already shown that discovery can be computationally hard even when the underlying optimization problem is easy. Our results show that this phenomenon persists, and becomes especially rich, in the two-graph setting. We obtain a detailed complexity picture, identifying tractable cases as well as strong hardness results.
Read the full abstract

All authors Hanno von Bergen, Larissa Fastenau, Enna Gerhard, Nicola Lorenz, Stephanie Maaz, Amer E. Mouawad, Roman Rabinovich, Nicole Schirrmacher, Daniel Schmand, Sebastian Siebertz, Mai Trinh

Mathematical Physics · Cross-listed from hep-th

TBA equations for $SU(r+1)$ quantum Seiberg-Witten curve: higher-order Mathieu equation

We develop the ODE/IM correspondence for the higher-order Mathieu equation arising from the quantum Seiberg-Witten curve of the pure $SU(r+1)$ ${\cal N}=2$ supersymmetric Yang-Mills theory. From the subdominant solutions, we construct the Q-/Y-systems and derive the corresponding TBA equations. The dependence of the moduli parameters is found to be encoded in the boundary conditions of the Y-functions at $\theta \to -\infty$. From these boundary data, we derive an analytic expression for the effective central charge, which also governs the subleading contribution in the large-$\theta$ expansion of the TBA equations. Finally, we compare the large-$\theta$ expansion of the Q-function derived from the TBA equations with that obtained from the WKB method, which yields analytic agreement at subleading order and precise numerical agreement at the higher-order corrections.
Read the full abstract

Probability

Expected hyperbolic volumes of random beta polytopes

Let $X_1,\ldots,X_n$ be independent random points in the closed unit ball of $\mathbb{R}^d$. Assume that each $X_i$ has a beta distribution with parameter $\beta_i \ge -1$: if $\beta_i>-1$, then $X_i$ has Lebesgue density proportional to $(1-\|x\|^2)^{\beta_i}$ on $\{\|x\|<1\}$, whereas the case $\beta_i=-1$ corresponds to the uniform distribution on the unit sphere $\{\|x\|=1\}$. Let $[X_1,\ldots,X_n]$ denote the convex hull of these points. Interpreting the unit ball as the Klein model of hyperbolic geometry, we derive closed-form formulas for the expected hyperbolic volume of the random hyperbolic polytope $[X_1,\ldots,X_n]$. As a special case, if $X_1,\ldots,X_n$ are independent and uniformly distributed on the unit sphere in $\mathbb{R}^3$, then for every $n\ge 4$, \[ \mathbb{E}\,\operatorname{Vol}_{3}^{\mathrm{hyp}}\!\bigl([X_1,\ldots,X_n]\bigr) = \pi\left(\frac{n}{2}-\sum_{j=1}^{n-1}\frac{1}{j}\right). \]
Read the full abstract

Logic · Cross-listed from cs.CC

Toward a Characterization of Simulation Between Arithmetic Theories

We study when a sound arithmetic theory $\mathcal S{\supseteq}S^1_2$ with polynomial-time decidable axioms has polynomial-size proofs of $Con_{\mathcal S{+}\phi}(n)$ for a true sentence $\phi$. Our structural result characterizes the nonexistence of a length-optimal propositional proof system: it is equivalent to every such $\mathcal S$ failing to simulate $S^1_2{+}\mathsf{Exp}{+}\phi_{BB}(k)$ for all sufficiently large $k$. Here $\phi_{BB}(k)$ asserts the exact $k$-state Busy Beaver value, and $\mathsf{Exp}$ asserts total exponentiation. Combining the Krajíček–Pudlák correspondence with our transfer theorem proves this equivalence. For fixed $\mathcal S$, a true extension not simulated by $\mathcal S$ exists if and only if eventual Busy Beaver non-simulation holds. A finite version transfers lower bounds to Busy Beaver instances with explicit parameter changes and proof-length bounds. For finitely axiomatized sequential $\mathcal S$, we prove that $S^1_2{\vdash}Con_{\mathcal S}{\rightarrow}Con_{\mathcal S{+}\phi}$ implies simulation. We refute the earlier conjecture that the corresponding $EA$ implication characterizes simulation. For these finitely axiomatized sequential theories, the Kolmogorov Blindness conjecture predicts exponential proof-length lower bounds for bounded consistency of true logarithmic-deficiency randomness extensions $x{\in}R^{\log}$, with an encoding-dependent rate and an onset common to all strings of each sufficiently large length. It predicts the same hardness after adjoining a true randomness axiom $y{\in}R^{\log}$ of equal length, provided $x$ remains random conditional on $y$. For fixed $\Pi_1$ axiom families with computable proof-length bounds and a common computable onset, we prove a quantitative equivalence between dense hardness with polynomially small exceptional fractions and hardness above a corresponding Kolmogorov-complexity threshold.
Read the full abstract

2604.27787 · 30 Apr · v4: Corrects relative consistency using $S^1_2$ and Busy Beaver transfer using $S^1_2{+}\mathsf{Exp}$; refutes the earlier Higher Relative Consistency conjecture; adds a Busy Beaver characterization of the nonexistence of length-optimal proof systems, finite transfer, and a quantitative density–randomness equivalence; consolidates the conjectures around Kolmogorov Blindness

Combinatorics

On the Extremal Energy of Complex Unit Gain Dumbbell Graphs

We study the extremal energy problem for complex unit gain graphs whose underlying graph is the dumbbell graph $D_{r,s,\ell}$. Using switching equivalence, we reduce the spectrum to the real parts of the two cycle gains and obtain an explicit expression of the characteristic polynomial in terms of matching polynomials of natural subgraphs. For the bipartite case, we determine the extremal gain assignments by coefficient comparison. For the non-bipartite cases, we analyze the Coulson integral kernels. Finally, the maximum-energy conditions are determined in all cases, while the minimum-energy conditions are determined except when $r$, $s$, and $\ell$ are all odd. For this remaining case, we alternatively prove sign restrictions for any improvement over $(0,0)$, and prove a Hessian criterion at the origin, which provides a sufficient condition for $(0,0)$ to fail to be an energy minimizer.
Read the full abstract

Symplectic Geometry

The existence criterion of holomorphic discs for higher $A_\infty$ operations via minimal discs

The main theorem of the paper provides an existence criterion of holomorphic discs for higher $A_\infty$ operations. The key step is to show that if a minimal disc in a Kähler manifold with boundary in a sequence of Lagrangian submanifolds intersecting transversely such that its partial Maslov indices are either all no less than $1$ or all no larger than $-1$, then there is a holomorphic disc with the same image as this minimal disc. As a by-product, we show that all minimal discs in $\C\mathrm{P}^m$ with boundary on $\R\mathrm{P}^m$ are holomorphic.
Read the full abstract

Combinatorics

Extension of Excess Demand Ascending Auction to Multi-Demand Model by Discrete Convex Analysis Approach

We consider the problem of finding the (unique) minimal Walrasian equilibrium price in multi-item, multi-unit auction models: there are multiple indivisible items for sale, with several units of each item, and a bidder may be interested in buying more than one copy of each item. In its special case with unit-demand bidders, where each bidder demands at most one unit of any item, Andersson, Andersson, and Talman (2013) proposed a general framework of ascending auction algorithms based on the concept of excess-demand item set. This paper extends this approach to the multi-unit case by exploiting the discrete convexity of the Lyapunov function associated with the auction model. In particular, we make use of the facts that (i) the equilibrium price vectors are characterized as the minimizers of the Lyapunov function, (ii) the Lyapunov function is an instance of an L-natural-convex function, and (iii) a concept generalizing “excess-demand item set” can be defined in L-natural-convex function minimization in general.
Read the full abstract

Functional Analysis

Iterated Aluthge transforms of some composition operators on weighted Bergman spaces

In this paper, we compute the iterated Aluthge transforms $\widetilde{C_\phi}^{(n)}$ of the composition operator $C_\phi$ on the weighted Bergman spaces $\mathcal{A}_\alpha^2(\mathbb{D})$, where $\phi(z)=az+(1-a)$ for $0<a<1$. Also, we obtain the norm and numerical radius of $\widetilde{C_\phi}^{(n)}$ on $\mathcal{A}_\alpha^2(\mathbb{D})$. We establish that $\widetilde{C_\phi}^{(n)}$ converges in the strong operator topology on $\mathcal{A}_\alpha^2(\mathbb{D})$. The purpose of this paper is to examine the results of \cite{jung2015iterated} for the weighted Bergman spaces $\mathcal{A}_\alpha^2(\mathbb{D})$. Additionally, by using the iterated Aluthge transforms of $C_\phi^*$ on $\mathcal{A}_\alpha^2(\mathbb{D})$, we derive the iterated Aluthge transforms of $C_\sigma$, where $\displaystyle\sigma(z)=\frac{az}{-(1-a)z+1}$ for $0<a<1$, on some weighted Hardy space $H^2(\beta_\alpha)$ and study its convergence. Finally, we raise some questions that emerge from these findings.
Read the full abstract

Numerical Analysis

Möbius-transformed trapezoidal rule for polynomial weights

This work studies numerical integration by the Möbius-transformed trapezoidal rule, which combines the classical trapezoidal rule with a change of variables induced by a Möbius transformation that maps the unit circle onto the real line. It is shown that this method achieves the optimal convergence rate for a polynomially weighted integral over the real line if the integrand lives in a related polynomially weighted Sobolev space with positive integer smoothness index. This result can also be generalized in a slightly weaker form for fractional smoothness indices via complex interpolation of function spaces. The algorithm only requires pointwise evaluations of the weight and the target integrand at prescribed nodes that do not depend on the integrand and weight in question. The established theoretical convergence rates are verified by numerical experiments.
Read the full abstract

Combinatorics

Gårding Polynomials

We introduce Gårding polynomials, a class of real multivariate polynomials characterized by positivity regions that are invariant under translation by positive vectors and closed under strictly positive affine transformations. We prove that this geometric formulation is equivalent both to a reduction to the multi-affine setting via polarization and to a recursive criterion in terms of partial derivatives. The class of Gårding polynomials strictly extends that of real stable polynomials while preserving many of their structural properties. In particular, multi-affine Gårding polynomials with nonnegative coefficients satisfy the Rayleigh property, and their positive univariate specializations have ultra log-concave coefficient sequences. The Gårding property for several matroid generating functions is preserved under natural matroid operations. As applications, we derive new negative dependence results for generating functions associated with various classes of matroids and graphs, including examples previously beyond the scope of real stability and Lorentzian methods. We further obtain analogous results for characteristic polynomials arising from certain matrix classes.
Read the full abstract

Information Theory

Why Self-Supervised Encoders Want to Be Normal

Self-supervised learning has achieved remarkable empirical success in learning robust representations without explicit labels, most recently demonstrated within the framework of Joint-Embedding Predictive Architectures (JEPA). However, a fundamental question remains: what analytical principles drive these encoders toward specific distributional states? In this paper, we demonstrate that the preference for normal distributions in self-supervised encoders is a direct consequence of the Information Bottleneck (IB) principle. By recasting the IB objective as a rate-distortion problem over the predictive manifold, we provide a theoretical basis for why optimal, target-neutral, latent representations should tend towards isotropic Gaussian states. Under this framework, we show that latent representations correspond to soft clustering of inputs sharing similar predictive distributions, organized within a natural simplex structure. This perspective unifies a wide range of existing supervised and less-supervised objectives and provides a principled explanation for commonly used regularization schemes. Furthermore, we derive practical loss objectives that approximate this structure and demonstrate their effectiveness on standard benchmarks. Ultimately, our framework offers a geometric lens to understanding representation collapse and it establishes a mathematical system for regularization strategies to be used to ensure high-entropy, informative embeddings in modern self-supervised models.
Read the full abstract

Analysis of PDEs

On the Lifespan of Axisymmetric Hall-MHD with Swirl

In this paper, we study the three-dimensional inviscid incompressible resistive Hall-MHD system in the axisymmetric setting with nontrivial swirl velocity and purely azimuthal magnetic. Assuming only that the swirl component of the initial velocity is sufficiently small, we prove that the lifespan of the strong solution can be sufficiently large. An explicit lifespan lower bound in terms of the size of the initial swirl is given. Moreover, we also study the behavior of the lifespan as the resistivity tends to zero.
Read the full abstract

Numerical Analysis

h-Adaptive FV Subcell Shock-Capturing for DGSEM on Heterogeneous Curvilinear Meshes

High-order methods offer superior dispersion and dissipation properties compared to low-order schemes but require robust stabilization for discontinuities. To ensure stability, local artificial viscosity is common, but often degrades sub-element resolution. Conversely, subcell resolution preserving limiting strategies such as the finite volume subcell method are typically restricted to uniform topologies, such as purely hexahedral or simplex meshes, or linear elements. This leaves a significant gap in treating the hybrid-element topologies necessary for complex engineering geometries. To bridge this gap, we introduce a robust shock-capturing approach for the discontinuous Galerkin spectral element method on mixed curvilinear meshes containing hexahedral, prismatic, tetrahedral, and pyramid elements. Non-hexahedral elements are handled via collapsed coordinate transformations. The proposed method utilizes an $h$-adaptive finite volume subcell scheme with an arbitrary subcell resolution up to $2\mathcal{N}+1$. Special care is taken to ensure conforming subcell distributions across different element types. Crucially, discrete conservation is proven theoretically and verified numerically, alongside validations for high-order convergence and robust shock capturing. Finally, the method's applicability to complex configurations is demonstrated through a simulation of the flow around a NACA 0012 airfoil.
Read the full abstract

All authors Anna Schwarz, Jens Keim, Christian Rohde, Andrea Beck

Numerical Analysis

An adaptive Deep Ritz framework for second-order fully nonlinear partial differential equations

As an alternative to PINNs, a Deep Ritz framework is proposed to solve fully nonlinear PDEs. A least-squares algorithm is advocated to decouple the nonlinearities from the variational features of several fully nonlinear PDEs. A splitting method allows to iteratively solve local nonlinear problems and linear variational problems at each iteration. While existing nonlinear solvers are applied to solve for nonlinearities, we propose a novel coupling with a Deep Ritz neural network approach that is well-suited to the variational flavor of the linear variational problems. An adaptive sampling strategy for the selection of collocation points is incorporated to increase the efficiency of the algorithm without sacrificing its accuracy. Numerical experiments are presented to solve the Dirichlet problem for several fully nonlinear equations, starting with the prototypical Monge-Ampère equation, showing the flexibility of the approach. Numerical results are compared with results obtained using a full PINNs approach. Finally, numerical experiments are extended to address the optimal transport Monge-Ampère problem with transport boundary conditions.
Read the full abstract

Symplectic Geometry

Inscriptions of Isosceles Trapezoids in Jordan Curves

We construct a Lagrangian Floer homology whose chain complex is generically generated by the inscriptions of isosceles trapezoids in a smooth Jordan curve. This is an extension of Greene and Lobb's Jordan Floer homology (arXiv:2404.05179), which we also call Jordan Floer homology. Its non-triviality re-establishes that every smooth Jordan curve inscribes every isosceles trapezoid. By consideration of the spectral invariants associated with the real filtration known as the action filtration, we establish new cases of non-smooth Jordan curves which admit inscriptions of isosceles trapezoids.
Read the full abstract

Optimization and Control · Cross-listed from eess.SY

Robust Geometric Control of Catenary Robots under Unstructured Force Uncertainties

This paper considers the robust control of a catenary robot composed of two quadrotors connected by an inextensible cable. The system is modeled on \(SE(3)\), with the cable treated as a geometric subsystem induced by the UAV configuration rather than as an independent dynamical element. The catenary shape determines configuration-dependent forces that couple the translational dynamics of the vehicles. We propose a geometric tracking controller for the relative configuration of the agents and analyze its robustness with respect to unstructured uncertainties in the catenary-induced forces. The main theoretical result establishes local input-to-state stability of the closed-loop tracking errors. In particular, we obtain asymptotic convergence in the nominal case and an explicit ultimate bound for the tracking errors under bounded catenary-force perturbations.
Read the full abstract