Mathematical Physics · Cross-listed from physics.flu-dyn
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Monday 16 February
Fluidic Shaping is a novel method for fabrication of optical components based on the equilibrium state of liquid volumes in neutral buoyancy, subjected to geometrical constraints. The underlying physics of this method is described by a highly nonlinear partial differential equation with Dirichlet boundary conditions and an integral constraint. To date, useful solutions for such optical liquid surfaces could be obtained analytically only for the linearized equations and only on circular or elliptical domains. A numerical solution for the non-linear equation was suggested, but only for the axi-symmetric case. Such solutions are, however, insufficient as they do not capture the full range of optical surfaces. Arbitrary domains offer an important degree of freedom for creating complex optical surfaces, and the nonlinear terms are essential for high quality solutions. Moreover, in the context of optics, it is not sufficient to resolve the shape of the surface, and it is essential to obtain accurate solutions for its curvature, which governs its optical properties. We here present the theoretical foundation for the Fluidic Shaping method over arbitrary domains, and the development of a high order (quintic) finite element numerical solver, capable of accurately resolving the topography and curvature of liquid interfaces on arbitrary domains. The code is based on reduced quintic finite elements, which we have modified to capture curved boundaries. We compare the results against low order finite elements and non-deformed high order elements, demonstrating the importance of high order approximations of both the solution and the domain. We also show the usability of the code for the prediction of optical surfaces derived from complex boundary conditions.
Combinatorics · Cross-listed from cs.LG
A Pragmatic Method for Comparing Clusterings with Overlaps and Outliers
Clustering algorithms are an essential part of the unsupervised data science ecosystem, and extrinsic evaluation of clustering algorithms requires a method for comparing the detected clustering to a ground truth clustering. In a general setting, the detected and ground truth clusterings may have outliers (objects belonging to no cluster), overlapping clusters (objects may belong to more than one cluster), or both, but methods for comparing these clusterings are currently undeveloped. In this note, we define a pragmatic similarity measure for comparing clusterings with overlaps and outliers, show that it has several desirable properties, and experimentally confirm that it is not subject to several common biases afflicting other clustering comparison measures.
Read the full abstract
Numerical Analysis
Domain decomposition dynamical low-rank for multi-dimensional radiative transfer equations
In this paper, we propose a domain decomposition dynamical low-rank method to solve high-dimensional radiative transfer problems and similar kinetic equations. The algorithm uses a separate low-rank approximation on each spatial subdomain, which means that, for a given accuracy, we can often use a smaller overall rank compared to classic dynamical low-rank methods. In particular, we can solve problems with point sources efficiently, that for classic algorithms require almost full rank. Our algorithm only transfers boundary data between subdomains and is thus very attractive for distributed memory parallelization, where classic dynamical low-rank algorithms suffer from global data dependency. We demonstrate the efficiency of our algorithm by a number of challenging test examples that have both very optical thin and thick regions.
Read the full abstract
Numerical Analysis · Cross-listed from cs.LG
BEACONS: Bounded-Error, Algebraically-Composable Neural Solvers for Partial Differential Equations
The traditional limitations of neural networks in reliably generalizing beyond the convex hulls of their training data present a significant problem for computational physics, in which one often wishes to solve PDEs in regimes far beyond anything which can be experimentally or analytically validated. In this paper, we show how it is possible to circumvent these limitations by constructing formally-verified neural network solvers for PDEs, with rigorous convergence, stability, and conservation properties, whose correctness can therefore be guaranteed even in extrapolatory regimes. By using the method of characteristics to predict the analytical properties of PDE solutions a priori (even in regions arbitrarily far from the training domain), we show how it is possible to construct rigorous extrapolatory bounds on the worst-case L^inf errors of shallow neural network approximations. Then, by decomposing PDE solutions into compositions of simpler functions, we show how it is possible to compose these shallow neural networks together to form deep architectures, based on ideas from compositional deep learning, in which the large L^inf errors in the approximations have been suppressed. The resulting framework, called BEACONS (Bounded-Error, Algebraically-COmposable Neural Solvers), comprises both an automatic code-generator for the neural solvers themselves, as well as a bespoke automated theorem-proving system for producing machine-checkable certificates of correctness. We apply the framework to a variety of linear and non-linear PDEs, including the linear advection and inviscid Burgers' equations, as well as the full compressible Euler equations, in both 1D and 2D, and illustrate how BEACONS architectures are able to extrapolate solutions far beyond the training data in a reliable and bounded way. Various advantages of the approach over the classical PINN approach are discussed.
Read the full abstract
Functional Analysis
Lower Estimates for $L_1$-Distortion of Transportation Cost Spaces
Quantifying the degree of dissimilarity between two probability distributions on a finite metric space is a fundamental task in Computer Science and Computer Vision. A natural dissimilarity measure based on optimal transport is the Earth Mover's Distance (EMD). A key technique for analyzing this metric, pioneered by Charikar (2002) and Indyk and Thaper (2003), involves constructing low-distortion embeddings of EMD(X) into the Lebesgue space $L_1$. It became a key problem to investigate whether the upper bound of $O(\log n)$ can be improved for important classes of metric spaces known to admit low-distortion embeddings into $L_1$. In the context of Computer Vision, grid graphs, especially planar grids, are among the most fundamental. Indyk posed the related problem of estimating the $L_1$-distortion of the space of uniform distributions on $n$-point subsets of $R^2$. The Progress Report, last updated in August 2011, highlighted two key results: first, the work of Khot and Naor (2006) on Hamming cubes, which showed that the $L_1$-distortion for Hamming cubes meets the described above upper estimate, and second, the result of Naor and Schechtman (2007) for planar grids, which established that the $L_1$-distortion of for a planar $n$ by $n$ grid is $\Omega(\sqrt{\log n})$. Our first result is the improvement of the lower bound on the $L_1$-distortion for grids to $\Omega(\log n)$, matching the universal upper bound up to multiplicative constants. The key ingredient allowing us to obtain these sharp estimates is a new Sobolev-type inequality for scalar-valued functions on the grid graphs. Our method is also applicable to many recursive families of graphs, such as diamond and Laakso graphs. We obtain the sharp distortion estimates of $\log n$ in these cases as well.
Read the full abstract
Algebraic Geometry
Calabi-Yau complete intersections associated to good pairs of generalized nef partitions
We introduce the notion of good pair of generalized nef partitions to describe Calabi-Yau complete intersections in Q-Fano toric varieties whose equations do not necessarily have maximal Newton polytopes. Moreover, we define a natural duality between them which generalizes Batyrev-Borisov mirror duality and allows to define a generalization of Berglund-Hübsch-Krawitz duality to quasismooth complete intersections.
Read the full abstract
Analysis of PDEs
Existence of large-data solutions to a thermo-piezoelectric system and forward operator analysis for associated inverse problems
We consider an inverse problem governed by the initial-boundary value problem for the thermo-piezoelectric Kelvin-Voigt dynamical system \[ \left\{ \begin{aligned} \rho(z,t) u_{tt} &= \frac{d}{dz} \big( \Gamma(\Theta) u_{zt} +p_1 u_z +p_2(z,t)\phi_z^0 +p_2(z,t)\chi_z -\beta \Theta \big), \\[1ex] 0 &=-\frac{d}{dz} \big( p_2(z,t)u_z -p_3(z,t)\phi_z^0 -p_3(z,t)\chi_z\big), \\ b(z,t)\Theta_t &= \frac{d}{dz}(k(z,t)\Theta_z) +\Gamma(\Theta)u_{zt}^2 -\beta \Theta u_{zt}. \end{aligned} \right. \] in an open bounded interval $\Omega\subset\mathbb{R}$, for the evolution of the displacement variable $u$, the electric potential $\phi^0$ and the temperature $\Theta\geq 0$, where $\chi$ is a given Dirichlet lift function. Assuming that the coefficients $\beta, p_1 \in \mathbb{R}^+$ and the the parameter functions $\rho$, $\Gamma$, $p_2$, $p_3$, $b$ and $k$ are strictly positive and bounded, a global-in-time existence result is established for weak solutions. We show that this can be achieved under energy- and entropy-minimal assumptions, in the sense that global weak solutions are shown to exist for any initial data $$u_0\in W^{1,2}(\Omega)\mbox{ with }u_0|_{\partial\Omega}\in\mathbb{R},\quad u_{0t}\in L^2(\Omega)\mbox{ with }u_{0t}|_{\partial\Omega}\in\mathbb{R}\quad\mbox{and}\quad 0\leq\Theta_0\in L^2(\Omega).$$ The qualitative analysis of the evolution problem then allows to model and analyze the structural properties of the corresponding forward operator arising in inverse parameter identification. Therein, two modeling approaches of the observation operator as approximations of the electrical surface charge are presented and results on their well-definedness and boundedness are established. Building on these results, we prove well-definedness, boundedness, and continuous Fr'{e}chet differentiability of the forward operator.
Read the full abstract
Metric Geometry
John-type decompositions for affinely-optimal positions of convex bodies
Many classical problems in convex geometry can be cast as optimization problems under certain containment conditions. The arguably best-understood example is volume-maximization of convex bodies contained in other convex bodies, where the John decomposition describes$\unicode{x2014}$and in the Euclidean case fully characterizes$\unicode{x2014}$the optimal positions. For many other such problems, however, no general optimality conditions are known. To address this, we generalize an approach of O. B. Ader to obtain a John-type decomposition as a necessary condition for affinely-optimal containment chains, i.e., chains $r L_1 + c \subseteq K \subseteq R L_2 + d$ for convex bodies $K, L_1, L_2 \subseteq \mathbb{R}^n$, translation vectors $c,d \in \mathbb{R}^n$, and reals $r,R > 0$ such that the ratio $\frac{R}{r}$ cannot be decreased by linearly transforming $K$. We again obtain sufficiency for optimality when ellipsoids are involved, and show how optimality conditions for various problems follow from our result. Our main applications concern the Banach-Mazur distance, where we provide necessary optimality conditions in the general case and a full characterization in the Euclidean case. Finally, we derive several consequences of these optimality conditions related to the Banach-Mazur distance to the Euclidean ball.
Read the full abstract
Representation Theory
Relative Character Asymptotics Beyond Stability for $\mathrm{PGL}_2 \times \mathrm{GL}_1$
The asymptotics of relative characters for real Lie groups were studied for representations $(\pi, \sigma)$ arising from Gan-Gross-Prasad pairs $(G,H)$ by Nelson and Venkatesh. They successfully compute the asymptotics of relative characters whenever the conductor of the associated Rankin-Selberg $L$-function $L(\pi \boxtimes \sigma^\vee)$ lies in a stable locus, i.e. away from conductor dropping. In this paper, we express asymptotics for relative characters in the non-archimedean setting for $(\mathrm{PGL}_2, \mathrm{GL}_1)$. The key new innovation is that our method overcomes the stability hypothesis and allows for significant conductor dropping.
Read the full abstract
Probability
Convergence for linear quadratic potential mean field games
This paper studies the limits of empirical means of open-loop Nash equilibria of linear-quadratic stochastic differential games as the number of players goes to infinity, when the corresponding mean field game is of potential type and may have multiple equilibria. Via weak compactness arguments, the limit points are characterized as optimal trajectories of the related deterministic control problem, thus ruling out some of the mean field equilibria. Our result is obtained by first connecting the finite player game to a suitable control problem, whose optimal trajectories are the empirical means of Nash equilibria of the game, and in which the number of players $N$ becomes a parameter. True convergence to the unique minimizer of the limit control problem then holds for almost every initial mean. In cases of multiple optimizers, we focus on examples to show that some symmetry of the data ensures that the sequence admits a random limit which is distributes uniformly among the minimizers of the potential. Multidimensional examples of the convergence result appear here for the first time, which show the flexibility of our method. We also establish a similar convergence results for the corresponding linear-quadratic potential mean field games with common noise, as the noise vanishes.
Read the full abstract
Information Theory
Constructions of linear codes from vectorial plateaued functions and their subfield codes with applications to quantum CSS codes
Linear codes over finite fields parameterized by functions have proven to be a powerful tool in coding theory, yielding optimal and few-weight codes with significant applications in secret sharing, authentication codes, and association schemes. In 2023, Xu et al. introduced a construction framework for 3-dimensional linear codes parameterized by two functions, which has demonstrated considerable success in generating infinite families of optimal linear codes. Motivated by this approach, we propose a construction that extends the framework to three functions, thereby enhancing the flexibility of the parameters. Additionally, we introduce a vectorial setting by allowing vector-valued functions, expanding the construction space and the set of achievable structural properties. We analyze both scalar and vectorial frameworks, employing Bent and s-Plateaued functions, including Almost Bent, to define the code generators. By exploiting the properties of the Walsh transform, we determine the explicit parameters and weight distributions of these codes and their punctured versions. A key result of this study is that the constructed codes have few weights, and their duals are distance and dimensionally optimal with respect to both the Sphere Packing and Griesmer bounds. Furthermore, we establish a theoretical connection between our vectorial approach and the classical first generic construction of linear codes, providing sufficient conditions for the resulting codes to be minimal and self-orthogonal. Finally, we investigate applications to quantum coding theory within the Calderbank-Shor-Steane framework.
Read the full abstract
Optimization and Control
On Convergence Analysis of Network-GIANT: An approximate Hessian-based fully distributed optimization algorithm
This paper presents a detailed convergence and performance analysis of a recently developed approximate Newton-type fully distributed optimization method for \(L\)-smooth, \(\mu\)-strongly convex local loss functions, called Network-GIANT (inspired by the Federated learning algorithm GIANT possessing mixed linear-quadratic convergence properties). Network-GIANT has been empirically seen to achieve faster linear convergence properties compared to its gradient-based counterparts, and several other existing second order distributed algorithms, while having the same communication complexity (per iteration) as its first order distributed counterparts. We first explicitly characterize a global linear convergence rate for Network-GIANT, which can be computed as the spectral radius of a $3 \times 3$ matrix dependent on $L$, $\mu$, and the spectral norm ($\sigma$) of the consensus matrix of the underlying undirected graph. We provide an explicit bound on the step size parameter $\eta$, below which this spectral radius is guaranteed to be less than $1$. Furthermore, we derive a mixed linear-quadratic inequality based upper bound for the optimality gap norm, and provide a rigorous proof of a local asymptotic convergence rate of \(1 - \eta \big(1 - \frac{\gamma}{\mu}\big)\) given the Hessian approximation error $\gamma < \mu$, which formally explains the faster convergence rate of Network-GIANT. Numerical experiments are carried out with a reduced CovType dataset for binary logistic regression over a variety of graphs, including heterogeneous data distributions, to illustrate the above theoretical results.
Read the full abstract
Complex Variables
Hodge theory for twisted log-differential forms
In this survey, we review recent developments in extending Hodge theory to differential forms with values in bundles equipped with singular metrics, based on joint work with Ya Deng, Christopher D. Hacon, and Mihai Păun.
Combinatorics
Diagonals in Riordan matrices and applications
We introduce a method for describing Riordan matrices via recurrence relations along their diagonals. This provides a new structural description that complements the classical row-wise and column-wise constructions via the A-sequence. As an application, we characterize families of palindromic Riordan polynomials, yielding a new combinatorial interpretation in terms of lattice paths.
Read the full abstract
Differential Geometry
On pp-waves with lightlike parallel spinors
We parametrize pp-wave spacetimes with compact codimension 2 hypersurfaces. In the vacuum case, we show that these spacetimes are locally in one-to-one correspondence with smooth curves of Riemannian Ricci-flat metrics modulo smooth curves of diffeomorphisms. We also prove that this one-to-one correspondence extends to pp-waves with prescribed null Ricci curvature. Moreover, the pp-wave spacetime carries a lightlike parallel spinor if and only if one (and hence all) of the Ricci-flat metrics carries a parallel spinor.
Read the full abstract
Optimization and Control
Effective approximations of solutions to highly oscillatory diffusion equations from coarse measurements
We approximate a diffusion equation with highly oscillatory coefficients with a diffusion equation with constant coefficients. The approach is put in action in contexts where only partial information (namely the global energy stored in the physical system) is available. While the reconstruction of the microstructure is known to be an ill-posed problem, we show that the reconstruction of effective coefficients is possible and this even with only some coarse information. The strategy we present takes the form of a non-convex optimization problem. Homogenization theory provides elements for a rigorous foundation of the approach. Some algorithmic aspects are discussed in details. We provide a comprehensive set of numerical illustrations that demonstrate the practical interest of our strategy. The present work improves on the earlier works [C. Le Bris, F. Legoll and S. Lemaire, COCV 2018; C. Le Bris, F. Legoll and K. Li, CRAS 2013].
Read the full abstract
Probability
Peaks over Threshold in Scale-Free Random Graphs
We explore extreme value phenomena in spatial scale-free random graphs in a continuum setting based on a homogeneous Poisson point process in $\mathbb{R}^d$. Vertices carry i.i.d. weights $(W_x)$ and, conditionally on the vertex set and the weights, edges are present independently with probability $p_{xy}=1-\exp\{-\lambda W_xW_y/|x-y|^\alpha\}$. Assuming Pareto-type weight tails with index $\beta>0$ and working in parameter ranges where degrees are almost surely finite, we study extremes and peaks over thresholds (POT) of edge lengths in a growing observation window. Our focus is the precise impact of the presence of (large) weights on edge lengths, captured through explicit scaling regimes and conditional POT limit theorems. Our main results identify a three-phase behavior governed by the weight-tail parameter $\beta$. We both deduce Fréchet-type limits for the maximum edge length itself and we reveal POT structures under a hub conditioning by proving a POT limit theorem. In the finite-mean regime $\beta>1$, the leading scaling agrees with the unweighted model up to a constant. By contrast, for $\beta\le 1$ the weights have a macroscopic effect on extreme edge lengths: for $\beta< 1$ the scaling changes, and the borderline case $\beta=1$ exhibits additional logarithmic corrections. The proofs combine Stein-type Poisson approximation via a Palm-coupling approach with a refined treatment of the dependence created by the conditioning event.
Read the full abstract
Operator Algebras
Continuous functions over a pure C*-algebra
Let $X$ be a compact metric space, and let $A$ be a pure $\mathrm{C}^*$-algebra. We show that $C(X,A)$ is pure whenever $A$ is simple; or every quotient of $A$ is stably finite (e.g., $A$ has stable rank one). Using permanence properties of pureness, we prove that the tensor product of any such $A$ with any ASH-algebra is pure.
Read the full abstract
Differential Geometry
Minimizing the volume of globally hyperbolic anti-de Sitter 3-manifolds
In this paper we show that the volume of a maximal globally hyperbolic Cauchy-compact anti-de Sitter $3$-manifold $M$ is at least $\pi^2|\chi(M)|$, and that this minimum value is attained if and only if $M$ is Fuchsian.
Information Theory
Center-Fed Pinching Antenna System (C-PASS): Modeling, Analysis, and Beamforming Design
A generalized framework for the novel center-fed pinching antenna system (C-PASS) is proposed. Within this framework, closed-form expressions for the degree of freedom (DoF) and power scaling law of the proposed C-PASS are first derived. These theoretical results reveal that the achievable DoF scales linearly with the number of input ports, $M$, and the number of receive antennas, $K$. Furthermore, the derived power scaling laws demonstrate that the C-PASS achieves a power gain of order $\mathcal{O}(P_T M)$, where $P_T$ denotes the transmit power. Based on the proposed C-PASS modeling, a sum-rate maximization problem for the joint optimization of transmit and pinching beamforming is then formulated. To solve this highly coupled non-convex problem, an efficient alternating optimization algorithm is developed. More particularly, the transmit precoding and power splitting ratios are updated via derived closed-form solutions, while the pinching antenna positions and radiation coefficients are optimized using block coordinate descent (BCD) methods. Finally, our numerical results reveal that the single-waveguide C-PASS: 1) achieves superior DoF and power scaling laws compared to the single-waveguide PASS; and 2) outperforms the multi-waveguide PASS in high-attenuation regimes, yielding a substantial gain exceeding $10$ dB.
Read the full abstract
Combinatorics
A mirror deformation of Markov numbers
We introduce a deformed squared Markov equation given by $X^2 + Y^2 + Z^2 + (q+q^{-1})(XY+YZ+XZ) = 3(1 + q + q^{-1})XYZ$. Symmetric solutions of this new equation present a remarkable factorization property which allows us to talk about their square roots. These square roots give a natural $q$-deformation of the Markov numbers that has not previously occurred in the literature. We call them mirror Markov numbers. We prove a characterization of mirror Markov numbers and discover a mutation rule, mirror mutation, to generate them all. We also prove a geometric realization of the corresponding mirror mutation on a once-punctured sphere with three orbifold points. Our mirror deformation leads to deformations of Fibonacci and Pell branches for which we give precise formulas. Furthermore, the deformed squared Markov equation specializes to many other very well known generalized Markov equations. We also obtain the super Markov numbers from a specialization of the deformed squared Markov numbers, which we use to prove a conjecture of Musiker.
Read the full abstract
Complex Variables
Identifying Bergman space functions from intervals
We characterize functions of a Bergman space on a square by their values and derivatives on the diagonals. This problem is connected with the reachable space of the one-dimensional heat equation on a finite interval with boundary $L^2$-controls.
Analysis of PDEs
On the Hölder continuity of signed solutions to doubly nonlinear parabolic equations in the mixed degenerate/singular cases
We prove the Hölder continuity of sign-changing solutions to the equation of the type $$\frac{\partial}{\partial t}\big(|u|^{q-1} u\big)- div\Big(|D u|^{p-2}\,D u\Big)=0,$$ where numbers $p$, $q$ satisfy the conditions $$0<q<p-1\quad \text{and}\quad p<2,$$ or $$q>p-1\quad\text{and}\quad p>2.$$ Our proof uses new versions of the integral Harnack type inequalities for sign-changing solutions.
Read the full abstract
Algebraic Geometry
On the behavior of analytic torsion for twisted canonical bundles under degenerations
Consider a degeneration of projective algebraic manifolds equipped with a compact group action over a curve. Suppose that the total space carries a Nakano semi-positive vector bundle, which is equivariant with respect to this action. We consider the relative canonical bundle twisted by this bundle. Under this setting, we prove that the logarithm of the equivariant analytic torsion of the regular fibers for this coefficient admits an asymptotic expansion near the discriminant locus. The leading term is given by a logarithmic singularity, while the subdominant term is given by a loglog-type singularity. In the non-equivariant case, we provide a formula for the coefficient of the leading term in terms of an integral of characteristic classes associated with the semi-stable reduction of the family. To establish these results, we prove the existence of an asymptotic expansion for both the equivariant Quillen metrics and the $L^{2}$-metrics in the above setting. We also calculate the leading term of the fiber integral of the Bott-Chern classes associated with the degeneration.
Read the full abstract
Algebraic Geometry
Non-quasi-$F$-split canonical affine fourfolds exist in every characteristic
We construct canonical $\mathbb{Q}$-factorial Gorenstein affine fourfolds in every positive characteristic that are not quasi-$F$-split.
Mathematical Physics · Cross-listed from cond-mat.mes-hall
Exact Multi-Valley Envelope Function Theory of Valley Splitting in Si/SiGe Nanostructures
Valley splitting in strained Si/SiGe quantum wells is a central parameter for silicon spin qubits and is commonly described with envelope-function and effective-mass theories. These models provide a computationally efficient continuum description and have been shown to agree well with atomistic approaches when the confinement potential is slowly varying on the lattice scale. In modern Si/SiGe heterostructures with atomically sharp interfaces and engineered Ge concentration profiles, however, the slowly varying potential approximation underlying conventional (local) envelope-function theory is challenged. We formulate an exact multi-valley envelope-function model by combining Burt-Foreman-type envelope-function theory, which does not rely on the assumption of a slowly varying potential, with a valley-sector decomposition of the Brillouin zone. This construction enforces band-limited envelopes, which satisfy a set of coupled integro-differential equations with a non-local potential energy operator. Using degenerate perturbation theory, we derive the intervalley coupling matrix element within this non-local model and prove that it is strictly invariant under global shifts of the confinement potential (choice of reference energy). We then show that the conventional local envelope model generically violates this invariance due to spectral leakage between valley sectors, leading to an unphysical energy-reference dependence of the intervalley coupling. The resulting ambiguity is quantified by numerical simulations of various engineered Si/SiGe heterostructures. Finally, we propose a simple spectrally filtered local approximation that restores the energy-reference invariance exactly and provides a good approximation to the exact non-local theory.
Read the full abstract
Numerical Analysis
A randomized global GMRES method for matrix equations
In this paper, we develop a new Randomized Global Generalized Minimum Residual (RGlGMRES) algorithm for efficiently computing solutions to large scale linear systems with multiple right hand sides. The proposed method builds on a recently developed randomized global Gram Schmidt process, in which sketched Frobenius inner products are employed to approximate the exact Frobenius inner products of high dimensional matrices. We give some new convergence results of the RGlGMRES method for multiple linear systems. In the case where the coefficient matrix A is diagonalizable, we derive new upper bounds for the randomized Frobenius norm of the residual. In this paper, we study how to introduce matrix sketching in this algorithm. It allows us to reduce the dimension of the problem in one of the main steps of the algorithm. We also show how to apply the RGlGMRES method for solving the Lyapunov matrix equation. To validate the effectiveness and practicality of this approach, we conduct several numerical experiments, which demonstrate that our RGlGMRES method is competitive with the global GMRES method for solving large scale multiple linear systems and Lyapunov matrix equation.
Read the full abstract
Optimization and Control
Reciprocal Specific Relative Entropy between Continuous Martingales
We introduce a novel notion of divergence between continuous martingales; the reciprocal specific relative entropy. First, we motivate this definition from multiple perspectives. Thereafter, we solve the reciprocal specific relative entropy minimization problem over the set of win-martingales (used as models for prediction markets Aldous (2013)). Surprisingly, we show that the optimizer is the renowned neutral Wright-Fisher diffusion. We also justify that this diffusion is in a sense the most salient win-martingale, since it is uniquely selected when we suitably perturb the degenerate martingale optimal transport problem of variance minimization.
Read the full abstract
Optimization and Control · Cross-listed from eess.SY
Hierarchical parameter estimation for distributed networked systems: a dynamic consensus approach
This work introduces a novel two-stage distributed framework to globally estimate constant parameters in a networked system, separating shared information from local estimation. The first stage uses dynamic average consensus to aggregate agents' measurements into surrogates of centralized data. Using these surrogates, the second stage implements a local estimator to determine the parameters. By designing an appropriate consensus gain, the persistence of excitation of the regressor matrix is achieved, and thus, exponential convergence of a local Gradient Estimator (GE) is guaranteed. The framework facilitates its extension to switched network topologies, quantization, and the heterogeneous substitution of the GE with a Dynamic Regressor Extension and Mixing (DREM) estimator, which supports relaxed excitation requirements.
Read the full abstract
Optimization and Control
An Age-Structured Vaccination Strategy for Epidemic Containment: A Model Predictive Control Approach
This work presents a novel Model Predictive Control (MPC) approach to develop an optimal age-structured vaccination strategy for the containment of COVID-19 in Wallonia, Belgium. The proposed MPC framework is designed to minimize deaths, achieve early disease eradication, and adhere to operational constraints. By incorporating an age-structured Susceptible-Infected-Recovered-Deceased (SIRD) model with an additional term for vaccination, the MPC strategy dynamically adapts to the evolving epidemic state. A detailed proof of the asymptotic stability and recursive feasibility of the proposed MPC algorithm is provided. This ensures that the optimal cost at each step provides an upper bound on the minimal number obtainable of deaths at the end of the pandemic. Moreover, simulations demonstrate that the proposed MPC approach outperforms the decreasing age vaccination strategy adopted by the Belgian government during the first wave of vaccinations. The results highlight the potential of MPC-based vaccination strategies to reduce the total number of deaths, accelerate disease eradication, and optimize vaccine administration.