Showing 1–33 of 33 results for author: Peña, M

The Stationary Behavior of Reflecting Coupled Brownian Motions with Applications to Shortest Remaining Processing Time Queues

Authors: Sixian Jin, Marvin Pena, Amber L. Puha

Abstract: With the objective of characterizing the stationary behavior of the scaling limit for shortest remaining processing time (SRPT) queues with a heavy-tailed processing time distribution, as obtained in Banerjee, Budhiraja, and Puha (BBP, 2022), we study reflecting coupled Brownian motions (RCBM) $(W_t(a), a, t \geq 0)$. These RCBM arise by regulating coupled Brownian motions (CBM)… ▽ More With the objective of characterizing the stationary behavior of the scaling limit for shortest remaining processing time (SRPT) queues with a heavy-tailed processing time distribution, as obtained in Banerjee, Budhiraja, and Puha (BBP, 2022), we study reflecting coupled Brownian motions (RCBM) $(W_t(a), a, t \geq 0)$. These RCBM arise by regulating coupled Brownian motions (CBM) $(χ_t(a), a,t \geq 0)$ to remain nonnegative. Here, for $t\geq 0$, $χ_t(0)=0$ and $χ_t(a):=w(a)+σB_t-μ(a)t$ for $a>0$, $w(\cdot)$ is a suitable initial condition, $σ$ is a positive constant, $B$ is a standard Brownian motion, and $μ(\cdot)$ is an unbounded, positive, strictly decreasing drift function. In the context of the BBP (2022) scaling limit, the drift function is determined by the model parameters, and, for each $a\geq 0$, $W_{\cdot}(a)$ represents the scaling limit of the amount of work in the system of size $a$ or less. Thus, for the BBP (2022) scaling limit, the time $t$ values of the RCBM describe the random distribution of the size of the remaining work in the system at time $t$. Our principal results characterize the stationary distribution of the RCBM in terms of a maximum process $M_*(\cdot)$ associated with CBM starting from zero. We obtain an explicit representation for the finite-dimensional distributions of $M_*(\cdot)$ and a simple formula for its covariance. We further show that the RCBM converge in distribution to $M_*(\cdot)$ as time $t$ approaches infinity. From this, we deduce the stationary behavior of the BBP (2022) scaling limit, including obtaining an integral expression for the stationary queue length in terms of the associated maximum process. While its distribution appears somewhat complex, we compute the mean and variance explicitly, and we connect with the work of Lin, Wierman, and Zwart (2011) to offer an illustration of Little's Law. △ Less

Submitted 3 July, 2025; originally announced July 2025.

Comments: 35 pages, 1 figure

MSC Class: Primary 60K25; 60F17; Secondary 60G57; 60G60; 68M20

On Bott's residue formula for toric complete intersections

Authors: Miguel Rodríguez Peña

Abstract: We determine the number of singularities - counted whit multiplicities - of generic distributions of dimension and codimension one on smooth complete intersections in compact toric orbifolds with isolated singularities. We also present some applications of this results. First, we analyze the case of regular distributions. As a second application, we establish a Poincaré-type inequality that relate… ▽ More We determine the number of singularities - counted whit multiplicities - of generic distributions of dimension and codimension one on smooth complete intersections in compact toric orbifolds with isolated singularities. We also present some applications of this results. First, we analyze the case of regular distributions. As a second application, we establish a Poincaré-type inequality that relates the multidegree of a foliation to the multidegrees of an invariant smooth complete intersection curve. △ Less

Submitted 24 June, 2025; v1 submitted 24 June, 2025; originally announced June 2025.

Coalitional model predictive control of an irrigation canal

Authors: Filiberto Fele, José M. Maestre, Mehdi Hashemy Shahdany, David Muñoz de la Peña, Eduardo F. Camacho

Abstract: We present a hierarchical control scheme for large-scale systems whose components can exchange information through a data network. The main goal of the supervisory layer is to find the best compromise between control performance and communicational costs by actively modifying the network topology. The actions taken at the supervisory layer alter the control agents' knowledge of the complete system… ▽ More We present a hierarchical control scheme for large-scale systems whose components can exchange information through a data network. The main goal of the supervisory layer is to find the best compromise between control performance and communicational costs by actively modifying the network topology. The actions taken at the supervisory layer alter the control agents' knowledge of the complete system, and the set of agents with which they can communicate. Each group of linked subsystems, or coalition, is independently controlled based on a decentralized model predictive control (MPC) scheme, managed at the bottom layer. Hard constraints on the inputs are imposed, while soft constraints on the states are considered to avoid feasibility issues. The performance of the proposed control scheme is validated on a model of the Dez irrigation canal, implemented on the accurate simulator for water systems SOBEK. Finally, the results are compared with those obtained using a centralized MPC controller. △ Less

Submitted 29 January, 2025; originally announced January 2025.

Comments: Single column version, 24 pages

MSC Class: 93-10; 93B45; 93A14 (Primary) 93-08; 49N10 (Secondary)

Journal ref: Journal of Process Control 24 (2014) 314 - 325

Processing the 2D and 3D Fresnel experimental databases via topological derivative methods

Authors: A. Carpio, M. Pena, M. L. Rapún

Abstract: This paper presents reconstructions of homogeneous targets from the 2D and 3D Fresnel databases by one-step imaging methods based on the computation of topological derivative and topological energy fields. The electromagnetic inverse scattering problem is recast as a constrained optimization problem, in which we seek to minimize the error when comparing experimental microwave measurements with com… ▽ More This paper presents reconstructions of homogeneous targets from the 2D and 3D Fresnel databases by one-step imaging methods based on the computation of topological derivative and topological energy fields. The electromagnetic inverse scattering problem is recast as a constrained optimization problem, in which we seek to minimize the error when comparing experimental microwave measurements with computer-generated synthetic data for arbitrary targets by approximating a Maxwell forward model. The true targets are then characterized by combining the topological derivatives or energies of such shape functionals for all available receivers and emitters at different frequencies. Our approximations are comparable to the best approximations already obtained by other methods. However, these topological fields admit easy to evaluate closed-form expressions, which speeds up the process. △ Less

Submitted 25 January, 2025; originally announced January 2025.

Journal ref: Inverse Problems 37, 105012, 2021

An evaluation algorithm for q-Bézier triangular patches formed by convex combinations

Authors: Jorge Delgado, Héctor Orera, Juan Manuel Peña

Abstract: An extension to triangular domains of the univariate q-Bernstein basis functions is introduced and analyzed. Some recurrence relations and properties such as partition of unity and degree elevation are proved for them. It is also proved that they form a basis for the space of polynomials of total degree less than or equal to n on a triangle. In addition, it is presented a de Casteljau type evaluat… ▽ More An extension to triangular domains of the univariate q-Bernstein basis functions is introduced and analyzed. Some recurrence relations and properties such as partition of unity and degree elevation are proved for them. It is also proved that they form a basis for the space of polynomials of total degree less than or equal to n on a triangle. In addition, it is presented a de Casteljau type evaluation algorithm whose steps are all linear convex combinations. △ Less

Submitted 22 January, 2025; originally announced January 2025.

Journal ref: J. Comput. Appl. Math. 428 (2023), Paper No. 115184, 11 pp

Accurate Bidiagonal Decomposition and Computations with Generalized Pascal Matrices

Authors: Jorge Delgado, Héctor Orera, Juan Manuel Peña

Abstract: This paper provides an accurate method to obtain the bidiagonal factorization of many generalized Pascal matrices, which in turn can be used to compute with high relative accuracy the eigenvalues, singular values and inverses of these matrices. Numerical examples are included. This paper provides an accurate method to obtain the bidiagonal factorization of many generalized Pascal matrices, which in turn can be used to compute with high relative accuracy the eigenvalues, singular values and inverses of these matrices. Numerical examples are included. △ Less

Submitted 21 January, 2025; originally announced January 2025.

MSC Class: 65F05; 65F15; 65G50; 15A23; 05A05; 11B65

Journal ref: Comput. Appl. Math. 391 (2021), Paper No. 113443, 10 pp

Optimal properties of tensor product of B-bases

Authors: Jorge Delgado, Héctor Orera, Juan Manuel Peña

Abstract: It is proved the optimal conditioning for the infinity norm of collocation matrices of the tensor product of normalized B-bases among the tensor product of all normalized totally positive bases of the corresponding space of functions. Bounds for the minimal eigenvalue and singular value and illustrative numerical examples are also included. It is proved the optimal conditioning for the infinity norm of collocation matrices of the tensor product of normalized B-bases among the tensor product of all normalized totally positive bases of the corresponding space of functions. Bounds for the minimal eigenvalue and singular value and illustrative numerical examples are also included. △ Less

Submitted 20 January, 2025; originally announced January 2025.

MSC Class: 65D17; 65F35; 15A123; 15A18

Journal ref: Appl. Math. Lett. 121 (2021), Paper No. 107473, 5 pp

Accurate algorithms for Bessel matrices

Authors: Jorge Delgado, Héctor Orera, Juan Manuel Peña

Abstract: In this paper, we prove that any collocation matrix of Bessel polynomials at positive points is strictly totally positive, that is, all its minors are positive. Moreover, an accurate method to construct the bidiagonal factorization of these matrices is obtained and used to compute with high relative accuracy the eigenvalues, singular values and inverses. Similar results for the collocation matrice… ▽ More In this paper, we prove that any collocation matrix of Bessel polynomials at positive points is strictly totally positive, that is, all its minors are positive. Moreover, an accurate method to construct the bidiagonal factorization of these matrices is obtained and used to compute with high relative accuracy the eigenvalues, singular values and inverses. Similar results for the collocation matrices for the reverse Bessel polynomials are also obtained. Numerical examples illustrating the theoretical results are included. △ Less

Submitted 17 January, 2025; originally announced January 2025.

MSC Class: 65F05; 65F15; 65G50; 33C10; 33C45; 15A23

Journal ref: J. Sci. Comput. 80 (2019), no. 2, 1264-1278

Infinity norm bounds for the inverse of Nekrasov matrices using scaling matrices

Authors: Héctor Orera, Juan Manuel Peña

Abstract: For many applications, it is convenient to have good upper bounds for the norm of the inverse of a given matrix. In this paper, we obtain such bounds when A is a Nekrasov matrix, by means of a scaling matrix transforming A into a strictly diagonally dominant matrix. Numerical examples and comparisons with other bounds are included. The scaling matrices are also used to derive new error bounds for… ▽ More For many applications, it is convenient to have good upper bounds for the norm of the inverse of a given matrix. In this paper, we obtain such bounds when A is a Nekrasov matrix, by means of a scaling matrix transforming A into a strictly diagonally dominant matrix. Numerical examples and comparisons with other bounds are included. The scaling matrices are also used to derive new error bounds for the linear complementarity problems when the involved matrix is a Nekrasov matrix. These error bounds can improve considerably other previous bounds. △ Less

Submitted 16 January, 2025; originally announced January 2025.

MSC Class: 65F35; 15A60; 65F05; 90C33

Journal ref: Appl. Math. Comput. 358 (2019), 119-127

Trefftz Discontinuous Galerkin approximation of an acoustic waveguide

Authors: Peter Monk, Manuel Pena, Virginia Selgas

Abstract: We propose a modified Trefftz Discontinuous Galerkin (TDG) method for approximating a time-harmonic acoustic scattering problem in an infinitely elongated waveguide. In the waveguide we suppose there is a bounded, penetrable and possibly absorbing scatterer. The classical TDG is not applicable to this important case. Novel features of our modified TDG method are that it is applicable when the scat… ▽ More We propose a modified Trefftz Discontinuous Galerkin (TDG) method for approximating a time-harmonic acoustic scattering problem in an infinitely elongated waveguide. In the waveguide we suppose there is a bounded, penetrable and possibly absorbing scatterer. The classical TDG is not applicable to this important case. Novel features of our modified TDG method are that it is applicable when the scatterer is absorbing, and it uses a stable treatment of the asymptotic radiation condition for the scattered field. For the modified TDG, we prove $h$ and $p$-convergence in the $L^2$ norm. The theoretical results are verified numerically for a discretization based on plane waves (that may be evanescent in the scatterer). △ Less

Submitted 14 May, 2025; v1 submitted 27 August, 2024; originally announced August 2024.

MSC Class: 65N15; 65N30; 35J05; 78A50

Combined matrices of almost strictly sign regular matrices

Authors: Pedro Alonso, Juan Manuel Peña, María Luisa Serrano

Abstract: The combined matrix is a very useful concept for many applications. Almost strictly sign regular (ASSR) matrices form an important structured class of matrices with two possible zero patterns, which are either type-I staircase or type-II staircase. We prove that, under an irreducibility condition, the pattern of zero and nonzero entries of an ASSR matrix is preserved by the corresponding combined… ▽ More The combined matrix is a very useful concept for many applications. Almost strictly sign regular (ASSR) matrices form an important structured class of matrices with two possible zero patterns, which are either type-I staircase or type-II staircase. We prove that, under an irreducibility condition, the pattern of zero and nonzero entries of an ASSR matrix is preserved by the corresponding combined matrix. Without the irreducibility condition, it is proved that type-I and type-II staircases are still preserved. Illustrative numerical examples are included. △ Less

Submitted 2 February, 2024; originally announced February 2024.

Comments: 8 pages

MSC Class: 65F05; 65F15; 65F35

Almost strictly sign regular rectangular matrices

Authors: P. Alonso, J. M. Peña, M. L. Serrano

Abstract: Almost strictly sign regular matrices are sign regular matrices with a special zero pattern and whose nontrivial minors are nonzero. In this paper we provide several properties of almost strictly sign regular rectangular matrices and analyze their QR factorization. Almost strictly sign regular matrices are sign regular matrices with a special zero pattern and whose nontrivial minors are nonzero. In this paper we provide several properties of almost strictly sign regular rectangular matrices and analyze their QR factorization. △ Less

Submitted 2 February, 2024; originally announced February 2024.

Comments: 10 pages

MSC Class: 65F05; 65F15; 65F35

Journal ref: Journal of Computational and Applied Mathematics 404 (2022) 113121

Inequalities and enumerative formulas for flags of Pfaff systems

Authors: Miguel Rodríguez Peña, Fernando Lourenço

Abstract: In this work, we study inequalities and enumerative formulas for flags of Pfaff systems on $\mathbb{P}^n_{\mathbb{C}}$. More specifically, we find the number of independent Pfaff systems that leave invariant a one-dimensional holomorphic foliation and deduce inequalities relating the degrees in the flags, which can be interpreted as the Poincaré problem for flags. Moreover, restricting to a flag o… ▽ More In this work, we study inequalities and enumerative formulas for flags of Pfaff systems on $\mathbb{P}^n_{\mathbb{C}}$. More specifically, we find the number of independent Pfaff systems that leave invariant a one-dimensional holomorphic foliation and deduce inequalities relating the degrees in the flags, which can be interpreted as the Poincaré problem for flags. Moreover, restricting to a flag of specific holomorphic foliations/distributions, we obtain inequalities involving the degrees. As a consequence, we prove stability results for the tangent sheaf of some rank two holomorphic foliations/distributions. △ Less

Submitted 30 January, 2024; originally announced January 2024.

On the Probability of Immunity

Authors: Jose M. Peña

Abstract: This work is devoted to the study of the probability of immunity, i.e. the effect occurs whether exposed or not. We derive necessary and sufficient conditions for non-immunity and $ε$-bounded immunity, i.e. the probability of immunity is zero and $ε$-bounded, respectively. The former allows us to estimate the probability of benefit (i.e., the effect occurs if and only if exposed) from a randomized… ▽ More This work is devoted to the study of the probability of immunity, i.e. the effect occurs whether exposed or not. We derive necessary and sufficient conditions for non-immunity and $ε$-bounded immunity, i.e. the probability of immunity is zero and $ε$-bounded, respectively. The former allows us to estimate the probability of benefit (i.e., the effect occurs if and only if exposed) from a randomized controlled trial, and the latter allows us to produce bounds of the probability of benefit that are tighter than the existing ones. We also introduce the concept of indirect immunity (i.e., through a mediator) and repeat our previous analysis for it. Finally, we propose a method for sensitivity analysis of the probability of immunity under unmeasured confounding. △ Less

Submitted 11 October, 2023; v1 submitted 21 September, 2023; originally announced September 2023.

Machine Learned Calabi-Yau Metrics and Curvature

Authors: Per Berglund, Giorgi Butbaia, Tristan Hübsch, Vishnu Jejjala, Damián Mayorga Peña, Challenger Mishra, Justin Tan

Abstract: Finding Ricci-flat (Calabi-Yau) metrics is a long standing problem in geometry with deep implications for string theory and phenomenology. A new attack on this problem uses neural networks to engineer approximations to the Calabi-Yau metric within a given Kähler class. In this paper we investigate numerical Ricci-flat metrics over smooth and singular K3 surfaces and Calabi-Yau threefolds. Using th… ▽ More Finding Ricci-flat (Calabi-Yau) metrics is a long standing problem in geometry with deep implications for string theory and phenomenology. A new attack on this problem uses neural networks to engineer approximations to the Calabi-Yau metric within a given Kähler class. In this paper we investigate numerical Ricci-flat metrics over smooth and singular K3 surfaces and Calabi-Yau threefolds. Using these Ricci-flat metric approximations for the Cefalú family of quartic twofolds and the Dwork family of quintic threefolds, we study characteristic forms on these geometries. We observe that the numerical stability of the numerically computed topological characteristic is heavily influenced by the choice of the neural network model, in particular, we briefly discuss a different neural network model, namely Spectral networks, which correctly approximate the topological characteristic of a Calabi-Yau. Using persistent homology, we show that high curvature regions of the manifolds form clusters near the singular points. For our neural network approximations, we observe a Bogomolov--Yau type inequality $3c_2 \geq c_1^2$ and observe an identity when our geometries have isolated $A_1$ type singularities. We sketch a proof that $χ(X~\smallsetminus~\mathrm{Sing}\,{X}) + 2~|\mathrm{Sing}\,{X}| = 24$ also holds for our numerical approximations. △ Less

Submitted 6 June, 2023; v1 submitted 17 November, 2022; originally announced November 2022.

Comments: Version accepted for publication: 48 pages, 32 figures, 8 tables, 3 appendices

Journal ref: ATMP v.27 no.4 (2023) 1107-1158

Bidiagonal Decompositions of Vandermonde-Type Matrices of Arbitrary Rank

Authors: Jorge Delgado, Plamen Koev, Ana Marco, Jose-Javier Martinez, Juan Manuel Pena, Per-Olof Persson, Steven Spasov

Abstract: We present a method to derive new explicit expressions for bidiagonal decompositions of Vandermonde and related matrices such as the (q-, h-) Bernstein-Vandermonde ones, among others. These results generalize the existing expressions for nonsingular matrices to matrices of arbitrary rank. For totally nonnegative matrices of the above classes, the new decompositions can be computed efficiently and… ▽ More We present a method to derive new explicit expressions for bidiagonal decompositions of Vandermonde and related matrices such as the (q-, h-) Bernstein-Vandermonde ones, among others. These results generalize the existing expressions for nonsingular matrices to matrices of arbitrary rank. For totally nonnegative matrices of the above classes, the new decompositions can be computed efficiently and to high relative accuracy componentwise in floating point arithmetic. In turn, matrix computations (e.g., eigenvalue computation) can also be performed efficiently and to high relative accuracy. △ Less

Submitted 18 October, 2022; originally announced October 2022.

MSC Class: 65F15; 15A23; 15B48; 15B35

On codimension one holomorphic distributions on compact toric orbifolds

Authors: Miguel Rodríguez Peña

Abstract: The number of singularities, counted with multiplicity, of a generic codimension one holomorphic distribution on a compact toric orbifold is determined. As a consequence, we give the classification of regular distributions on rational normal scrolls and weighted projective spaces. Under certain conditions, we also prove that the singular set of a codimension one holomorphic foliation on a toric or… ▽ More The number of singularities, counted with multiplicity, of a generic codimension one holomorphic distribution on a compact toric orbifold is determined. As a consequence, we give the classification of regular distributions on rational normal scrolls and weighted projective spaces. Under certain conditions, we also prove that the singular set of a codimension one holomorphic foliation on a toric orbifold admits at least one irreducible component of codimension two, and we give a Darboux-Jouanolou type integrability theorem for codimension one holomorphic foliations. We illustrate our results with several examples. △ Less

Submitted 26 February, 2024; v1 submitted 24 June, 2022; originally announced June 2022.

Comments: Accepted in International Journal of Mathematics

Neural Network Approximations for Calabi-Yau Metrics

Authors: Vishnu Jejjala, Damian Kaloni Mayorga Pena, Challenger Mishra

Abstract: Ricci flat metrics for Calabi-Yau threefolds are not known analytically. In this work, we employ techniques from machine learning to deduce numerical flat metrics for the Fermat quintic, for the Dwork quintic, and for the Tian-Yau manifold. This investigation employs a single neural network architecture that is capable of approximating Ricci flat Kaehler metrics for several Calabi-Yau manifolds of… ▽ More Ricci flat metrics for Calabi-Yau threefolds are not known analytically. In this work, we employ techniques from machine learning to deduce numerical flat metrics for the Fermat quintic, for the Dwork quintic, and for the Tian-Yau manifold. This investigation employs a single neural network architecture that is capable of approximating Ricci flat Kaehler metrics for several Calabi-Yau manifolds of dimensions two and three. We show that measures that assess the Ricci flatness of the geometry decrease after training by three orders of magnitude. This is corroborated on the validation set, where the improvement is more modest. Finally, we demonstrate that discrete symmetries of manifolds can be learned in the process of learning the metric. △ Less

Submitted 27 January, 2021; v1 submitted 31 December, 2020; originally announced December 2020.

Comments: v2: 42 pages, figures improved, discrete symmetries section added, discussions enhanced, references added

Towards Conditional Path Analysis

Authors: Jose M. Peña

Abstract: We extend path analysis by giving sufficient conditions for computing the partial covariance of two random variables from their covariance. This is specifically done by correcting the covariance with the product of some partial variance ratios. As a result, the partial covariance retains the covariance's salient feature of factorizing over the edges in the paths between the two variables of intere… ▽ More We extend path analysis by giving sufficient conditions for computing the partial covariance of two random variables from their covariance. This is specifically done by correcting the covariance with the product of some partial variance ratios. As a result, the partial covariance retains the covariance's salient feature of factorizing over the edges in the paths between the two variables of interest. △ Less

Submitted 21 December, 2020; originally announced December 2020.

Comments: arXiv admin note: substantial text overlap with arXiv:2002.05226

Localization formulas on complex supermanifolds

Authors: Leonardo Abath, Maurício Corrêa, Miguel Rodríguez Peña

Abstract: In this work we provide a localization formulae for odd holomorphic super vector fields on compact complex supermanifolds with fermionic dimension equal to the bosonic dimension. We prove a residue theorem for holomorphic super vector fields with non-degenerated and isolated singularities and we determine the residues under certain generic local conditions. In this work we provide a localization formulae for odd holomorphic super vector fields on compact complex supermanifolds with fermionic dimension equal to the bosonic dimension. We prove a residue theorem for holomorphic super vector fields with non-degenerated and isolated singularities and we determine the residues under certain generic local conditions. △ Less

Submitted 28 September, 2019; originally announced September 2019.

Comments: 40 pages

On the Poincaré problem for foliations on compact toric orbifolds

Authors: Miguel Rodríguez Peña

Abstract: We give an optimal upper bound of the degree of quasi-smooth hypersurfaces which are invariant by a one-dimensional holomorphic foliation on a compact toric orbifold, i.e. on a complete simplicial toric variety. This bound depends only on the degree of the foliation and of the degrees of the toric homogeneous coordinates. We give an optimal upper bound of the degree of quasi-smooth hypersurfaces which are invariant by a one-dimensional holomorphic foliation on a compact toric orbifold, i.e. on a complete simplicial toric variety. This bound depends only on the degree of the foliation and of the degrees of the toric homogeneous coordinates. △ Less

Submitted 6 September, 2021; v1 submitted 27 April, 2019; originally announced April 2019.

Comments: Accepted in Geometriae Dedicata

Extremal and optimal properties of B-bases Collocation Matrices

Authors: Jorge Delgado, J. M. Peña

Abstract: Totally positive matrices are related with the shape preserving representations of a space of functions. The normalized B-basis of the space has optimal shape preserving properties. B-splines and rational Bernstein bases are examples of normalized B-bases. Some results on the optimal conditioning and on extremal properties of the minimal eigenvalue and singular value of the collocation matrices of… ▽ More Totally positive matrices are related with the shape preserving representations of a space of functions. The normalized B-basis of the space has optimal shape preserving properties. B-splines and rational Bernstein bases are examples of normalized B-bases. Some results on the optimal conditioning and on extremal properties of the minimal eigenvalue and singular value of the collocation matrices of normalized B-bases are proved. Numerical examples confirm the theoretical results and answer related questions. △ Less

Submitted 28 January, 2019; originally announced January 2019.

Comments: 11 pages

MSC Class: 65F35; 65F15; 15B48; 15A12; 15A18; 65D17

SVD update methods for large matrices and applications

Authors: Juan Manuel Peña, Tomas Sauer

Abstract: We consider the problem of updating the SVD when augmenting a "tall thin" matrix, i.e., a rectangular matrix $A \in \RR^{m \times n}$ with $m \gg n$. Supposing that an SVD of $A$ is already known, and given a matrix $B \in \RR^{m \times n'}$, we derive an efficient method to compute and efficiently store the SVD of the augmented matrix $[ A B ] \in \RR^{m \times (n+n')}$. This is an important tool… ▽ More We consider the problem of updating the SVD when augmenting a "tall thin" matrix, i.e., a rectangular matrix $A \in \RR^{m \times n}$ with $m \gg n$. Supposing that an SVD of $A$ is already known, and given a matrix $B \in \RR^{m \times n'}$, we derive an efficient method to compute and efficiently store the SVD of the augmented matrix $[ A B ] \in \RR^{m \times (n+n')}$. This is an important tool for two types of applications: in the context of principal component analysis, the dominant left singular vectors provided by this decomposition form an orthonormal basis for the best linear subspace of a given dimension, while from the right singular vectors one can extract an orthonormal basis of the kernel of the matrix. We also describe two concrete applications of these concepts which motivated the development of our method and to which it is very well adapted. △ Less

Submitted 10 September, 2018; originally announced September 2018.

MSC Class: 65F30

Identification of Strong Edges in AMP Chain Graphs

Authors: Jose M. Peña

Abstract: The essential graph is a distinguished member of a Markov equivalence class of AMP chain graphs. However, the directed edges in the essential graph are not necessarily strong or invariant, i.e. they may not be shared by every member of the equivalence class. Likewise for the undirected edges. In this paper, we develop a procedure for identifying which edges in an essential graph are strong. We als… ▽ More The essential graph is a distinguished member of a Markov equivalence class of AMP chain graphs. However, the directed edges in the essential graph are not necessarily strong or invariant, i.e. they may not be shared by every member of the equivalence class. Likewise for the undirected edges. In this paper, we develop a procedure for identifying which edges in an essential graph are strong. We also show how this makes it possible to bound some causal effects when the true chain graph is unknown. △ Less

Submitted 25 June, 2018; v1 submitted 23 November, 2017; originally announced November 2017.

Comments: arXiv admin note: text overlap with arXiv:1303.0691

Journal ref: UAI 2018

Optimal interval length for the collocation of the Newton basis

Authors: J. M. Carnicer, Y. Khiar, J. M. Peña

Abstract: It is known that the Lagrange interpolation problem at equidistant nodes is ill-conditioned. We explore the influence of the interval length in the computation of divided differences of the Newton interpolation formula. Condition numbers are computed for lower triangular matrices associated to the Newton interpolation formula at equidistant nodes. We consider the collocation matrices $L$ and… ▽ More It is known that the Lagrange interpolation problem at equidistant nodes is ill-conditioned. We explore the influence of the interval length in the computation of divided differences of the Newton interpolation formula. Condition numbers are computed for lower triangular matrices associated to the Newton interpolation formula at equidistant nodes. We consider the collocation matrices $L$ and $P_L$ of the monic Newton basis and a normalized Newton basis, so that $P_L$ is the lower triangular Pascal matrix. In contrast to $L$, $P_L$ does not depend on the interval length, and we show that the Skeel condition number of the $(n+1)\times (n+1)$ lower triangular Pascal matrix is $3^n$. The $\infty$-norm condition number of the collocation matrix $L$ of the monic Newton basis is computed in terms of the interval length. The minimum asymptotic growth rate is achieved for intervals of length 3. △ Less

Submitted 20 September, 2017; originally announced September 2017.

MSC Class: 41A05; 65F35; 15A12

Decomposition of tensors

Authors: Juan Manuel Peña, Tomas Sauer

Abstract: We consider representations of tensors as sums of decomposable tensors or, equivalently, decomposition of multilinear forms into one--forms. In this short note we show that there exists a particular finite strongly orthogonal decomposition which is essentially unique and yields all critical points of the multilinear form on the torus. In particular, this determines exactly the number of critical p… ▽ More We consider representations of tensors as sums of decomposable tensors or, equivalently, decomposition of multilinear forms into one--forms. In this short note we show that there exists a particular finite strongly orthogonal decomposition which is essentially unique and yields all critical points of the multilinear form on the torus. In particular, this determines exactly the number of critical points of the multilinear form, giving an affirmative answer to a finiteness conjecture by Friedland. △ Less

Submitted 18 September, 2014; originally announced September 2014.

F-Theory on all Toric Hypersurface Fibrations and its Higgs Branches

Authors: Denis Klevers, Damian Kaloni Mayorga Pena, Paul-Konstantin Oehlmann, Hernan Piragua, Jonas Reuter

Abstract: We consider F-theory compactifications on genus-one fibered Calabi-Yau manifolds with their fibers realized as hypersurfaces in the toric varieties associated to the 16 reflexive 2D polyhedra. We present a base-independent analysis of the codimension one, two and three singularities of these fibrations. We use these geometric results to determine the gauge groups, matter representations, 6D matter… ▽ More We consider F-theory compactifications on genus-one fibered Calabi-Yau manifolds with their fibers realized as hypersurfaces in the toric varieties associated to the 16 reflexive 2D polyhedra. We present a base-independent analysis of the codimension one, two and three singularities of these fibrations. We use these geometric results to determine the gauge groups, matter representations, 6D matter multiplicities and 4D Yukawa couplings of the corresponding effective theories. All these theories have a non-trivial gauge group and matter content. We explore the network of Higgsings relating these theories. Such Higgsings geometrically correspond to extremal transitions induced by blow-ups in the 2D toric varieties. We recover the 6D effective theories of all 16 toric hypersurface fibrations by repeatedly Higgsing the theories that exhibit Mordell-Weil torsion. We find that the three Calabi-Yau manifolds without section, whose fibers are given by the toric hypersurfaces in P^2, P^1x P^1 and the recently studied P^2(1,1,2), yield F-theory realizations of SUGRA theories with discrete gauge groups Z_3, Z_2 and Z_4. This opens up a whole new arena for model building with discrete global symmetries in F-theory. In these three manifolds, we also find codimension two I_2-fibers supporting matter charged only under these discrete gauge groups. Their 6D matter multiplicities are computed employing ideal techniques and the associated Jacobian fibrations. We also show that the Jacobian of the biquadric fibration has one rational section, yielding one U(1)-gauge field in F-theory. Furthermore, the elliptically fibered Calabi-Yau manifold based on dP_1 has a U(1)-gauge field induced by a non-toric rational section. In this model, we find the first F-theory realization of matter with U(1)-charge q=3. △ Less

Submitted 2 September, 2014; v1 submitted 20 August, 2014; originally announced August 2014.

Comments: 129 pages, 32 figures, 44 tables v2: minor changes, references added

Learning AMP Chain Graphs under Faithfulness

Authors: Jose M. Peña

Abstract: This paper deals with chain graphs under the alternative Andersson-Madigan-Perlman (AMP) interpretation. In particular, we present a constraint based algorithm for learning an AMP chain graph a given probability distribution is faithful to. We also show that the extension of Meek's conjecture to AMP chain graphs does not hold, which compromises the development of efficient and correct score+search… ▽ More This paper deals with chain graphs under the alternative Andersson-Madigan-Perlman (AMP) interpretation. In particular, we present a constraint based algorithm for learning an AMP chain graph a given probability distribution is faithful to. We also show that the extension of Meek's conjecture to AMP chain graphs does not hold, which compromises the development of efficient and correct score+search learning algorithms under assumptions weaker than faithfulness. △ Less

Submitted 24 April, 2012; originally announced April 2012.

Unobservable Planar Bimodal Linear Systems: Miniversal Deformations, Controllability and Stabilization

Authors: Josep Ferrer, M. Dolors Magret, Juan R. Pacha, Marta Peña

Abstract: We consider the set of bimodal linear systems consisting of two linear dynamics acting on each side of a given hyperplane, assuming continuity along the separating hyperplane. Focusing on the unobservable planar ones, we obtain a simple explicit characterization of controllability. Moreover, we apply the canonical forms of these systems depending on two state variables to obtain explicitly miniver… ▽ More We consider the set of bimodal linear systems consisting of two linear dynamics acting on each side of a given hyperplane, assuming continuity along the separating hyperplane. Focusing on the unobservable planar ones, we obtain a simple explicit characterization of controllability. Moreover, we apply the canonical forms of these systems depending on two state variables to obtain explicitly miniversal deformations, to illustrate bifurcation diagrams and to prove that the unobservable controllable systems are stabilizable. Preprint of an article submitted for consideration in IJBC ©2011 copyright World Scientific Publishing Company http://www.worldscinet.com/ijbc/ △ Less

Submitted 30 December, 2011; originally announced January 2012.

Comments: 18 pages, 3 figures

Towards Optimal Learning of Chain Graphs

Authors: Jose M. Peña

Abstract: In this paper, we extend Meek's conjecture (Meek 1997) from directed and acyclic graphs to chain graphs, and prove that the extended conjecture is true. Specifically, we prove that if a chain graph H is an independence map of the independence model induced by another chain graph G, then (i) G can be transformed into H by a sequence of directed and undirected edge additions and feasible splits and… ▽ More In this paper, we extend Meek's conjecture (Meek 1997) from directed and acyclic graphs to chain graphs, and prove that the extended conjecture is true. Specifically, we prove that if a chain graph H is an independence map of the independence model induced by another chain graph G, then (i) G can be transformed into H by a sequence of directed and undirected edge additions and feasible splits and mergings, and (ii) after each operation in the sequence H remains an independence map of the independence model induced by G. Our result has the same important consequence for learning chain graphs from data as the proof of Meek's conjecture in (Chickering 2002) had for learning Bayesian networks from data: It makes it possible to develop efficient and asymptotically correct learning algorithms under mild assumptions. △ Less

Submitted 25 September, 2011; originally announced September 2011.

Finding Consensus Bayesian Network Structures

Authors: Jose M. Peña

Abstract: Suppose that multiple experts (or learning algorithms) provide us with alternative Bayesian network (BN) structures over a domain, and that we are interested in combining them into a single consensus BN structure. Specifically, we are interested in that the consensus BN structure only represents independences all the given BN structures agree upon and that it has as few parameters associated as po… ▽ More Suppose that multiple experts (or learning algorithms) provide us with alternative Bayesian network (BN) structures over a domain, and that we are interested in combining them into a single consensus BN structure. Specifically, we are interested in that the consensus BN structure only represents independences all the given BN structures agree upon and that it has as few parameters associated as possible. In this paper, we prove that there may exist several non-equivalent consensus BN structures and that finding one of them is NP-hard. Thus, we decide to resort to heuristics to find an approximated consensus BN structure. In this paper, we consider the heuristic proposed in \citep{MatzkevichandAbramson1992,MatzkevichandAbramson1993a,MatzkevichandAbramson1993b}. This heuristic builds upon two algorithms, called Methods A and B, for efficiently deriving the minimal directed independence map of a BN structure relative to a given node ordering. Methods A and B are claimed to be correct although no proof is provided (a proof is just sketched). In this paper, we show that Methods A and B are not correct and propose a correction of them. △ Less

Submitted 11 July, 2011; v1 submitted 10 January, 2011; originally announced January 2011.

Comments: Changes from v3 to v4: Section 1 has been extended with more motivation and a review of the literature. Theorem 3 proves that CONSENSUS is not only NP-hard but NP-complete. A flaw in Theorem 4 has been fixed. The proof of Theorem 5 has been re-written from scratch. Now, it is self-contained, i.e. it does not rely upon the algorithm by Chickering (2004)

Journal ref: Journal of Artificial Intelligence Research, 42, 661-687, 2011

Reading Dependencies from Covariance Graphs

Authors: Jose M. Peña

Abstract: The covariance graph (aka bi-directed graph) of a probability distribution $p$ is the undirected graph $G$ where two nodes are adjacent iff their corresponding random variables are marginally dependent in $p$. In this paper, we present a graphical criterion for reading dependencies from $G$, under the assumption that $p$ satisfies the graphoid properties as well as weak transitivity and compositio… ▽ More The covariance graph (aka bi-directed graph) of a probability distribution $p$ is the undirected graph $G$ where two nodes are adjacent iff their corresponding random variables are marginally dependent in $p$. In this paper, we present a graphical criterion for reading dependencies from $G$, under the assumption that $p$ satisfies the graphoid properties as well as weak transitivity and composition. We prove that the graphical criterion is sound and complete in certain sense. We argue that our assumptions are not too restrictive. For instance, all the regular Gaussian probability distributions satisfy them. △ Less

Submitted 26 June, 2012; v1 submitted 21 October, 2010; originally announced October 2010.

Comments: Changes from v1 to v2: Minor cosmetic changes, plus the addition of reference (Richardson and Spirtes, 2002) in page 8. Changes from v2 to v3: Addition of some references; International Journal of Approximate Reasoning, 2012

Faithfulness in Chain Graphs: The Gaussian Case

Authors: Jose M. Peña

Abstract: This paper deals with chain graphs under the classic Lauritzen-Wermuth-Frydenberg interpretation. We prove that the regular Gaussian distributions that factorize with respect to a chain graph $G$ with $d$ parameters have positive Lebesgue measure with respect to $\mathbb{R}^d$, whereas those that factorize with respect to $G$ but are not faithful to it have zero Lebesgue measure with respect to… ▽ More This paper deals with chain graphs under the classic Lauritzen-Wermuth-Frydenberg interpretation. We prove that the regular Gaussian distributions that factorize with respect to a chain graph $G$ with $d$ parameters have positive Lebesgue measure with respect to $\mathbb{R}^d$, whereas those that factorize with respect to $G$ but are not faithful to it have zero Lebesgue measure with respect to $\mathbb{R}^d$. This means that, in the measure-theoretic sense described, almost all the regular Gaussian distributions that factorize with respect to $G$ are faithful to it. △ Less

Submitted 13 August, 2010; originally announced August 2010.

MSC Class: 62H05; 60E05; 68T30

Journal ref: Proceedings of the 14th International Conference on Artificial Intelligence and Statistics (AISTATS 2011), 588-599