Pasar al contenido principal

PREDICCION, INFERENCIA Y COMPUTACION EN MODELOS ESTRUCTURADOS DE ALTA DIMENSION

PGC2018-101643-B-I00

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de Generación de Conocimiento y Fortalecimiento Científico y Tecnológico del Sistema de I+D+i
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D (Excelencia / Generación de Conocimiento)
Año convocatoria 2018
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POMPEU FABRA CCT

Resultados relacionados

Resultados totales (Incluyendo duplicados): 32
Encontrada(s) 1 página(s)

Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Non-existence results for stochastic wave equations in one dimension

  • Foondun, Mohammud
  • Nualart, Eulàlia
The purpose of this paper is to extend recent results of [2] and [10] for the stochastic heat equation to the stochastic wave equation given by [...] where Ẇ is space-time white noise, σ is a real-valued globally Lipschitz function but b is assumed to be only locally Lipschitz continuous. Three types of domain conditions are studied: D = [0, 1] with homogeneous Dirichlet boundary conditions, D = [0, 2π] with periodic boundary conditions, and D = R. Then, under suitable conditions, the following integrability condition [...] is studied in relation to non-existence of global solutions., EN acknowledges support from the Spanish MINECO grant PGC2018-101643-B-I00 and Ayudas Fundacion BBVA a Equipos de Investigación Científica 2017.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Instantaneous everywhere-blowup of parabolic SPDEs

  • Foondun, Mohammud
  • Khoshnevisan, Davar
  • Nualart, Eulàlia
Davar Khoshnevisan: Research supported in part by the United States’ National Science Foundation Grants DMS-1855439 and DMS-2245242. Eulalia Nualart: Acknowledges support from the Spanish MINECO Grant PGC2018-101643-B-I00 and Ayudas Fundacion BBVA a Proyectos de Investigación Científica 2021.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Density estimates for jump diffusion processes

  • Kohatsu-Higa, Arturo
  • Nualart, Eulàlia
  • Tran, Ngoc Khue
We consider a real-valued diffusion process with a linear jump term driven by a Poisson point process and we assume that the jump amplitudes have a centered density with finite moments. We show upper and lower estimates for the density of the solution in the case that the jump amplitudes follow a Gaussian or Laplacian law. The proof of the lower bound uses a general expression for the density of the solution in terms of the convolution of the density of the continuous part and the jump amplitude density. The upper bound uses an upper tail estimate in terms of the jump amplitude distribution and techniques of the Malliavin calculus in order to bound the density by the tails of the solution. We also extend the lower bounds to the multidimensional case., Eulalia Nualart acknowledges support from the Spanish MINECO grant PGC2018-101643-B-I00 and Ayudas Fundacion BBVA a Equipos de Investigación Científica 2017. Ngoc Khue Tran acknowledges support from the Vietnam Institute for Advanced Study in Mathematics (VIASM ) where a part of this work was done during his visit.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

The neurogenic fate of the hindbrain boundaries relies on Notch3-dependent asymmetric cell divisions

  • Hevia, Covadonga F.
  • Engel-Pizcueta, Carolyn
  • Udina, Frederic
  • Pujades Corbi, Cristina
Elucidating the cellular and molecular mechanisms that regulate the balance between progenitor cell proliferation and neuronal differentiation in the construction of the embryonic brain demands the combination of cell lineage and functional approaches. Here, we generate the comprehensive lineage of hindbrain boundary cells by using a CRISPR-based knockin zebrafish transgenic line that specifically labels the boundaries. We unveil that boundary cells asynchronously engage in neurogenesis undergoing a functional transition from neuroepithelial progenitors to radial glia cells, coinciding with the onset of Notch3 signaling that triggers their asymmetrical cell division. Upon notch3 loss of function, boundary cells lose radial glia properties and symmetrically divide undergoing neuronal differentiation. Finally, we show that the fate of boundary cells is to become neurons, the subtype of which relies on their axial position, suggesting that boundary cells contribute to refine the number and proportion of the distinct neuronal populations., This work was funded by grant PGC2018-095663-B-I00 from Spanish Ministry of Science and Innovation (MICIN), Agencia Estatal de Investigación (AEI), and Fondo Europeo de Desarrollo Regional (FEDER) to C.P. The Department of Medicine and Life Sciences (UPF) is an Unidad de Excelencia María de Maeztu funded by the MICIN and the AEI (DOI: 10.13039/501100011033) Ref: CEX2018-000792-M. C.E.P. is a recipient of a predoctoral FPU fellowship from the Spanish Ministry of Universities. F.U.’s work was supported by PGC2018-101643-B-I00 (MICIN/AEI-FEDER). C.P. is a recipient of ICREA Academia award (Generalitat de Catalunya).
Repositorio Digital de la UPF
Publicación Capítulo o parte de un libro (BookPart)

Mean estimation in high dimension

  • Lugosi, Gábor
In this note we discuss the statistical problem of estimating the mean of a random vector based on independent, identically distributed data. This classical problem has recently attracted a lot of attention both in mathematical statistics and in theoretical computer science and numerous intricacies have been revealed. We discuss some of the recent advances, focusing on high-dimensional aspects., This work was supported by the Spanish Ministry of Economy and Competitiveness, Grant PGC2018-101643-B-I00 and by “Google Focused Award Algorithms and Learning for AI”.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Concentration of the spectral norm of Erdös-Rényi random graphs

  • Lugosi, Gábor
  • Mendelson, Shahar
  • Zhivotovskiy, Nikita
We present results on the concentration properties of the spectral norm ||Ap|| of the adjacency matrix Ap of an Erdös-Rényi random graph G(n, p). First, we consider the Erdös-Rényi random graph process and prove that ||Ap|| is uniformly concentrated over the range p E [C log n/n, 1]. The analysis is based on delocalization arguments, uniform laws of large numbers, together with the entropy method to prove concentration inequalities. As an application of our techniques, we prove sharp sub-Gaussian moment inequalities for ||Ap|| for all p E [c log3 n/n, 1] that improve the general bounds of Alon, Krivelevich, and Vu (Israel J. Math. 131 (2002) 259-267) and some of the more recent results of Erdös et al. (Ann. Probab. 41 (2013) 2279-2375). Both results are consistent with the asymptotic result of Füredi and Komlós (Combinatorica 1 (1981) 233-241) that holds for fixed p as n -> infinite., G. Lugosi was supported by the Spanish Ministry of Economy and Competitiveness, Grant PGC2018-101643-B-I00; High-dimensional problems in structured probabilistic models - Ayudas Fundación BBVA a Equipos de Investigación Científica 2017; and Google Focused Award Algorithms and Learning for AI. S. Mendelson was supported in part by the Israel Science Foundation. N. Zhivotovskiy was supported by RSF Grant 18-11-00132. This work was done while N. Zhivotovskiy was a postdoctoral fellow at the Department of Mathematics, Technion I.I.T. and researcher at National University Higher School of Economics.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Concentration of posterior model probabilities and normalized L0 criteria

  • Rossell Ribera, David
We study frequentist properties of Bayesian and L0 model selection, with a focus on (potentially non-linear) high-dimensional regression. We propose a construction to study how posterior probabilities and normalized L0 criteria concentrate on the (Kullback-Leibler) optimal model and other subsets of the model space. When such concentration occurs, one also bounds the frequentist probabilities of selecting the correct model, type I and type II errors. These results hold generally, and help validate the use of posterior probabilities and L0 criteria to control frequentist error probabilities associated to model selection and hypothesis tests. Regarding regression, we help understand the effect of the sparsity imposed by the prior or the L0 penalty, and of problem characteristics such as the sample size, signal-to-noise, dimension and true sparsity. A particular finding is that one may use less sparse formulations than would be asymptotically optimal, but still attain consistency and often also significantly better finite-sample performance. We also prove new results related to misspecifying the mean or covariance structures, and give tighter rates for certain non-local priors than currently available., DR was partially funded by the Europa Excelencia grant EUR2020-112096, NIH grant R01 CA158113-01, Ramón y Cajal Fellowship RYC-2015-18544, Plan Estatal PGC2018-101643-B-I00 and Ayudas Fundación BBVA a equipos de investigación científica en Big Data 2017.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Optimal convergence rates for the invariant density estimation of jump-diffusion processes

  • Amorino, Chiara
  • Nualart, Eulàlia
We aim at estimating the invariant density associated to a stochastic differential equation with jumps in low dimension, which is for d = 1 and d = 2. We consider a class of fully non-linear jump diffusion processes whose invariant density belongs to some Hölder space. Firstly, in dimension one, we show that the kernel density estimator achieves the convergence rate 1/T, which is the optimal rate in the absence of jumps. This improves the convergence rate obtained in Amorino and Gloter [J. Stat. Plann. Inference 213 (2021) 106–129], which depends on the Blumenthal-Getoor index for d = 1 and is equal to (logT)/T for d = 2. Secondly, when the jump and diffusion coefficients are constant and the jumps are finite, we show that is not possible to find an estimator with faster rates of estimation. Indeed, we get some lower bounds with the same rates {1/T, (logT)/T} in the mono and bi-dimensional cases, respectively. Finally, we obtain the asymptotic normality of the estimator in the one-dimensional case for the fully non-linear process., CA gratefully acknowledges financial support of ERC Consolidator Grant 815703 “STAMFORD: Statistical Methods for High Dimensional Diffusions”. EN acknowledges support from the Spanish MINECO grant PGC2018-101643-B-I00 and Ayudas Fundación BBVA a Equipos de Investigación Científica 2017.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Secant varieties of toric varieties arising from simplicial complexes

  • Khadam, Muhammad Azeem
  • Michalek, Mateusz
  • Zwiernik, Piotr
Motivated by the study of the secant variety of the Segre-Veronese variety we propose a general framework to analyze properties of the secant varieties of toric embeddings of affine spaces defined by simplicial complexes. We prove that every such secant is toric, which gives a way to use combinatorial tools to study singularities. We focus on the Segre-Veronese variety for which we completely classify their secants that give Gorenstein or Q-Gorenstein varieties. We conclude providing the explicit description of the singular locus.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

General Bayesian loss function selection and the use of improper models

  • Jewson, Jack
  • Rossell Ribera, David
Statisticians often face the choice between using probability models or a paradigm defined by minimising a loss function. Both approaches are useful and, if the loss can be re-cast into a proper probability model, there are many tools to decide which model or loss is more appropriate for the observed data, in the sense of explaining the data's nature. However, when the loss leads to an improper model, there are no principled ways to guide this choice. We address this task by combining the Hyvärinen score, which naturally targets infinitesimal relative probabilities, and general Bayesian updating, which provides a unifying framework for inference on losses and models. Specifically we propose the ℋ-score, a general Bayesian selection criterion and prove that it consistently selects the (possibly improper) model closest to the data-generating truth in Fisher's divergence. We also prove that an associated ℋ-posterior consistently learns optimal hyper-parameters featuring in loss functions, including a challenging tempering parameter in generalised Bayesian inference. As salient examples, we consider robust regression and non-parametric density estimation where popular loss functions define improper models for the data and hence cannot be dealt with using standard model selection tools. These examples illustrate advantages in robustness-efficiency trade-offs and enable Bayesian inference for kernel density estimation, opening a new avenue for Bayesian non-parametrics., Jack Jewson and David Rossell were partially funded by the Ayudas Fundación BBVA a Equipos de Investigación Cientifica 2017 and Proyectos de Investigación Científica en Matemáticas 2021, and Government of Spain's Plan Nacional PGC2018-101643-B-I00 grants. David Rossell was also partially funded by the Europa Excelencia grant EUR2020-112096 and Ramón y Cajal Fellowship RYC-2015-18544.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Heterogeneous large datasets integration using Bayesian factor regression

  • Avalos-Pacheco, Alejandra
  • Rossell Ribera, David
  • Savage, Richard S.
Two key challenges in modern statistical applications are the large amount of information recorded per individual, and that such data are often not collected all at once but in batches. These batch effects can be complex, causing distortions in both mean and variance. We propose a novel sparse latent factor regression model to integrate such heterogeneous data. The model provides a tool for data exploration via dimensionality reduction and sparse low-rank covariance estimation while correcting for a range of batch effects. We study the use of several sparse priors (local and non-local) to learn the dimension of the latent factors. We provide a flexible methodology for sparse factor regression which is not limited to data with batch effects. Our model is fitted in a deterministic fashion by means of an EM algorithm for which we derive closed-form updates, contributing a novel scalable algorithm for non-local priors of interest beyond the immediate scope of this paper. We present several examples, with a focus on bioinformatics applications. Our results show an increase in the accuracy of the dimensionality reduction, with non-local priors substantially improving the reconstruction of factor cardinality. The results of our analyses illustrate how failing to properly account for batch effects can result in unreliable inference. Our model provides a novel approach to latent factor regression that balances sparsity with sensitivity in scenarios both with and without batch effects and is highly computationally efficient., Partially funded by the NIH grant R01 CA158113-01, RyC-2015-18544, Ayudas Fundación BBVA a equipos de investigación científica 2017 and Spanish Plan Estatal grant PGC2018-101643-B-I00.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Learning partial correlation graphs and graphical models by covariance queries

  • Lugosi, Gábor
  • Truszkowski, Jakub
  • Velona, Vasiliki
  • Zwiernik, Piotr
We study the problem of recovering the structure underlying large Gaussian graphical models or, more generally, partial correlation graphs. In high-dimensional problems it is often too costly to store the entire sample covariance matrix. We propose a new input model in which one can query single entries of the covariance matrix. We prove that it is possible to recover the support of the inverse covariance matrix with low query and computational complexity. Our algorithms work in a regime when this support is represented by tree-like graphs and, more generally, for graphs of small treewidth. Our results demonstrate that for large classes of graphs, the structure of the corresponding partial correlation graphs can be determined much faster than even computing the empirical covariance matrix., GL, VV, and PZ were supported by the Spanish Ministry of Economy and Competitiveness, Grant PGC2018-101643-B-I00 and FEDER, EU. GL and PZ acknowledge the support of "High-dimensional problems in structured probabilistic models - Ayudas Fundación BBVA a Equipos de Investigación Cientifica 2017". GL was supported by "Google Focused Award Algorithms and Learning for AI" and PZ by Beatriu de Pinós grant (BP-2016-00002) and Ramón y Cajal (RYC-2017-22544).
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Dependence in elliptical partial correlation graphs

  • Rossell Ribera, David
  • Zwiernik, Piotr
The Gaussian model equips strong properties that facilitate studying and interpreting graphical models. Specifically it reduces conditional independence and the study of positive association to determining partial correlations and their signs. When Gaussianity does not hold partial correlation graphs are a useful relaxation of graphical models, but it is not clear what information they contain (besides the obvious lack of linear association). We study elliptical and transelliptical distributions as middleground between the Gaussian and other families that are more flexible but either do not embed strong properties or do not lead to simple interpretation. We characterize the meaning of zero partial correlations in elliptical and elliptical copula models and show that they retain much of the dependence structure from the Gaussian case. Regarding positive dependence, we prove impossibility results to learn certain positive (trans)elliptical graphical models, including that an elliptical distribution that is multivariate totally positive of order two for all dimensions must be essentially Gaussian. We then show how to interpret positive partial correlations as a relaxation, and obtain important properties related to faithfulness and Simpson’s paradox. We illustrate the transelliptical model potential to study tail dependence in S&P500 data, and of positivity to improve regularized inference., DR and PZ were supported from the Spanish Government grants (RYC-2015-18544, RYC2017-22544, PGC2018-101643-B-I00), and Ayudas Fundación BBVA a Equipos de Investigación Cientifica 2017. DR was partially supported by NIH grant R01 CA158113-01 and Europa Excelencia EUR2020-112096.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Approximate Laplace approximations for scalable model selection

  • Rossell Ribera, David
  • Abril Pla, Oriol
  • Bhattacharya, Anirban
Includes supplementary materials for the online appendix., We propose the approximate Laplace approximation (ALA) to evaluate integrated likelihoods, a bottleneck in Bayesian model selection. The Laplace approximation (LA) is a popular tool that speeds up such computation and equips strong model selection properties. However, when the sample size is large or one considers many models the cost of the required optimizations becomes impractical. ALA reduces the cost to that of solving a least-squares problem for each model. Further, it enables efficient computation across models such as sharing pre-computed sufficient statistics and certain operations in matrix decompositions. We prove that in generalized (possibly non-linear) models ALA achieves a strong form of model selection consistency for a suitably-defined optimal model, at the same functional rates as exact computation. We consider fixed- and high-dimensional problems, group and hierarchical constraints, and the possibility that all models are misspecified. We also obtain ALA rates for Gaussian regression under non-local priors, an important example where the LA can be costly and does not consistently estimate the integrated likelihood. Our examples include non-linear regression, logistic, Poisson and survival models. We implement the methodology in the R package mombf., Spanish Government grants Europa Excelencia, Grant/Award Number: EUR2020-112096, RYC-2015-18544 and PGC2018-101643-B-I00; NIH, Grant/Award Number: R01 CA158113DMS-01.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Multivariate mean estimation with direction-dependent accuracy

  • Lugosi, Gábor
  • Mendelson, Shahar
We consider the problem of estimating the mean of a random vector based on N independent, identically distributed observations.We prove the existence of an estimator that has a nearoptimal error in all directions in which the variance of the one-dimensional marginal of the random vector is not too small: with probability 1−δ, the procedure returns μN which satisfies, for every direction u∈Sd−1, ⟨μN−μ,u⟩≤NC(σ(u)log(1/δ)+(E∥X−EX∥2)1/2), where σ2(u)=Var(⟨X,u⟩) and C is a constant. To achieve this, we require only slightly more than the existence of the covariance matrix, in the form of a certain moment-equivalence assumption., Gábor Lugosi was supported by the Spanish Ministry of Economy and Competitivity, Grant PGC2018-101643-B-I00 and FEDER, EU, and by 'Google Focused Award Algorithms and Learning for AI'.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

A dynamic and expandable digital 3D-atlas maker for monitoring the temporal changes in tissue growth during hindbrain morphogenesis

  • Blanc, Matthias
  • Dalmasso, Giovanni
  • Udina, Frederic
  • Pujades Corbi, Cristina
Reconstruction of prototypic three-dimensional (3D) atlases at the scale of whole tissues or organs requires specific methods to be developed. We have established a digital 3D-atlas maker (DAMAKER) and built a digital 3D-atlas to monitor the changes in the growth of the neuronal differentiation domain in the zebrafish hindbrain upon time. DAMAKER integrates spatial and temporal data of cell populations, neuronal differentiation and brain morphogenesis, through <i>in vivo</i> imaging techniques paired with image analyses and segmentation tools. First, we generated a 3D-reference from several imaged hindbrains and segmented them using a trainable tool; these were aligned using rigid registration, revealing distribution of neuronal differentiation growth patterns along the axes. Second, we quantified the dynamic growth of the neuronal differentiation domain by <i>in vivo</i> neuronal birthdating experiments. We generated digital neuronal birthdating 3D-maps and revealed that the temporal order of neuronal differentiation prefigured the spatial distribution of neurons in the tissue, with an inner-outer differentiation gradient. Last, we applied it to specific differentiated neuronal populations such as glutamatergic and GABAergic neurons, as proof-of-concept that the digital birthdating 3D-maps could be used as a proxy to infer neuronal birthdate. As this protocol uses open-access tools and algorithms, it can be shared for standardized, accessible, tissue-wide cell population atlas construction., This work was funded by grants PGC2018-095663-B-I00 and RED2018-102553-T to CP and PGC2018-101643-B-I00 to FU, from Spanish MCIN/AEI (DOI: 10.13039/501100011033) and Fondo Europeo de Desarrollo Regional (FEDER). Department of Medicine and Life Sciences UPF is a Unidad de Excelencia María de Maeztu funded by the Spanish MCIN/AEI (DOI: 10.13039/501100011033), Ref CEX2018-000792-M. CP is a recipient of ICREA Academia award (Generalitat de Catalunya).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Capítulo o parte de un libro (BookPart)

Mean estimation in high dimension

  • Lugosi, Gábor
In this note we discuss the statistical problem of estimating the mean of a random vector based on independent, identically distributed data. This classical problem has recently attracted a lot of attention both in mathematical statistics and in theoretical computer science and numerous intricacies have been revealed. We discuss some of the recent advances, focusing on high-dimensional aspects., This work was supported by the Spanish Ministry of Economy and Competitiveness, Grant PGC2018-101643-B-I00 and by “Google Focused Award Algorithms and Learning for AI”.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Learning partial correlation graphs and graphical models by covariance queries

  • Lugosi, Gábor
  • Truszkowski, Jakub
  • Velona, Vasiliki
  • Zwiernik, Piotr
We study the problem of recovering the structure underlying large Gaussian graphical models or, more generally, partial correlation graphs. In high-dimensional problems it is often too costly to store the entire sample covariance matrix. We propose a new input model in which one can query single entries of the covariance matrix. We prove that it is possible to recover the support of the inverse covariance matrix with low query and computational complexity. Our algorithms work in a regime when this support is represented by tree-like graphs and, more generally, for graphs of small treewidth. Our results demonstrate that for large classes of graphs, the structure of the corresponding partial correlation graphs can be determined much faster than even computing the empirical covariance matrix., GL, VV, and PZ were supported by the Spanish Ministry of Economy and Competitiveness, Grant PGC2018-101643-B-I00 and FEDER, EU. GL and PZ acknowledge the support of "High-dimensional problems in structured probabilistic models - Ayudas Fundación BBVA a Equipos de Investigación Cientifica 2017". GL was supported by "Google Focused Award Algorithms and Learning for AI" and PZ by Beatriu de Pinós grant (BP-2016-00002) and Ramón y Cajal (RYC-2017-22544).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Concentration of the spectral norm of Erdös-Rényi random graphs

  • Lugosi, Gábor
  • Mendelson, Shahar
  • Zhivotovskiy, Nikita
We present results on the concentration properties of the spectral norm ||Ap|| of the adjacency matrix Ap of an Erdös-Rényi random graph G(n, p). First, we consider the Erdös-Rényi random graph process and prove that ||Ap|| is uniformly concentrated over the range p E [C log n/n, 1]. The analysis is based on delocalization arguments, uniform laws of large numbers, together with the entropy method to prove concentration inequalities. As an application of our techniques, we prove sharp sub-Gaussian moment inequalities for ||Ap|| for all p E [c log3 n/n, 1] that improve the general bounds of Alon, Krivelevich, and Vu (Israel J. Math. 131 (2002) 259-267) and some of the more recent results of Erdös et al. (Ann. Probab. 41 (2013) 2279-2375). Both results are consistent with the asymptotic result of Füredi and Komlós (Combinatorica 1 (1981) 233-241) that holds for fixed p as n -> infinite., G. Lugosi was supported by the Spanish Ministry of Economy and Competitiveness, Grant PGC2018-101643-B-I00; High-dimensional problems in structured probabilistic models - Ayudas Fundación BBVA a Equipos de Investigación Científica 2017; and Google Focused Award Algorithms and Learning for AI. S. Mendelson was supported in part by the Israel Science Foundation. N. Zhivotovskiy was supported by RSF Grant 18-11-00132. This work was done while N. Zhivotovskiy was a postdoctoral fellow at the Department of Mathematics, Technion I.I.T. and researcher at National University Higher School of Economics.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Multivariate mean estimation with direction-dependent accuracy

  • Lugosi, Gábor
  • Mendelson, Shahar
We consider the problem of estimating the mean of a random vector based on N independent, identically distributed observations.We prove the existence of an estimator that has a nearoptimal error in all directions in which the variance of the one-dimensional marginal of the random vector is not too small: with probability 1−δ, the procedure returns μN which satisfies, for every direction u∈Sd−1, ⟨μN−μ,u⟩≤NC(σ(u)log(1/δ)+(E∥X−EX∥2)1/2), where σ2(u)=Var(⟨X,u⟩) and C is a constant. To achieve this, we require only slightly more than the existence of the covariance matrix, in the form of a certain moment-equivalence assumption., Gábor Lugosi was supported by the Spanish Ministry of Economy and Competitivity, Grant PGC2018-101643-B-I00 and FEDER, EU, and by 'Google Focused Award Algorithms and Learning for AI'.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Secant varieties of toric varieties arising from simplicial complexes

  • Khadam, Muhammad Azeem
  • Michalek, Mateusz
  • Zwiernik, Piotr
Motivated by the study of the secant variety of the Segre-Veronese variety we propose a general framework to analyze properties of the secant varieties of toric embeddings of affine spaces defined by simplicial complexes. We prove that every such secant is toric, which gives a way to use combinatorial tools to study singularities. We focus on the Segre-Veronese variety for which we completely classify their secants that give Gorenstein or Q-Gorenstein varieties. We conclude providing the explicit description of the singular locus.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A dynamic and expandable digital 3D-atlas maker for monitoring the temporal changes in tissue growth during hindbrain morphogenesis

  • Blanc, Matthias
  • Dalmasso, Giovanni
  • Udina, Frederic
  • Pujades Corbi, Cristina
Reconstruction of prototypic three-dimensional (3D) atlases at the scale of whole tissues or organs requires specific methods to be developed. We have established a digital 3D-atlas maker (DAMAKER) and built a digital 3D-atlas to monitor the changes in the growth of the neuronal differentiation domain in the zebrafish hindbrain upon time. DAMAKER integrates spatial and temporal data of cell populations, neuronal differentiation and brain morphogenesis, through <i>in vivo</i> imaging techniques paired with image analyses and segmentation tools. First, we generated a 3D-reference from several imaged hindbrains and segmented them using a trainable tool; these were aligned using rigid registration, revealing distribution of neuronal differentiation growth patterns along the axes. Second, we quantified the dynamic growth of the neuronal differentiation domain by <i>in vivo</i> neuronal birthdating experiments. We generated digital neuronal birthdating 3D-maps and revealed that the temporal order of neuronal differentiation prefigured the spatial distribution of neurons in the tissue, with an inner-outer differentiation gradient. Last, we applied it to specific differentiated neuronal populations such as glutamatergic and GABAergic neurons, as proof-of-concept that the digital birthdating 3D-maps could be used as a proxy to infer neuronal birthdate. As this protocol uses open-access tools and algorithms, it can be shared for standardized, accessible, tissue-wide cell population atlas construction., This work was funded by grants PGC2018-095663-B-I00 and RED2018-102553-T to CP and PGC2018-101643-B-I00 to FU, from Spanish MCIN/AEI (DOI: 10.13039/501100011033) and Fondo Europeo de Desarrollo Regional (FEDER). Department of Medicine and Life Sciences UPF is a Unidad de Excelencia María de Maeztu funded by the Spanish MCIN/AEI (DOI: 10.13039/501100011033), Ref CEX2018-000792-M. CP is a recipient of ICREA Academia award (Generalitat de Catalunya).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Concentration of posterior model probabilities and normalized L0 criteria

  • Rossell Ribera, David
We study frequentist properties of Bayesian and L0 model selection, with a focus on (potentially non-linear) high-dimensional regression. We propose a construction to study how posterior probabilities and normalized L0 criteria concentrate on the (Kullback-Leibler) optimal model and other subsets of the model space. When such concentration occurs, one also bounds the frequentist probabilities of selecting the correct model, type I and type II errors. These results hold generally, and help validate the use of posterior probabilities and L0 criteria to control frequentist error probabilities associated to model selection and hypothesis tests. Regarding regression, we help understand the effect of the sparsity imposed by the prior or the L0 penalty, and of problem characteristics such as the sample size, signal-to-noise, dimension and true sparsity. A particular finding is that one may use less sparse formulations than would be asymptotically optimal, but still attain consistency and often also significantly better finite-sample performance. We also prove new results related to misspecifying the mean or covariance structures, and give tighter rates for certain non-local priors than currently available., DR was partially funded by the Europa Excelencia grant EUR2020-112096, NIH grant R01 CA158113-01, Ramón y Cajal Fellowship RYC-2015-18544, Plan Estatal PGC2018-101643-B-I00 and Ayudas Fundación BBVA a equipos de investigación científica en Big Data 2017.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Dependence in elliptical partial correlation graphs

  • Rossell Ribera, David
  • Zwiernik, Piotr
The Gaussian model equips strong properties that facilitate studying and interpreting graphical models. Specifically it reduces conditional independence and the study of positive association to determining partial correlations and their signs. When Gaussianity does not hold partial correlation graphs are a useful relaxation of graphical models, but it is not clear what information they contain (besides the obvious lack of linear association). We study elliptical and transelliptical distributions as middleground between the Gaussian and other families that are more flexible but either do not embed strong properties or do not lead to simple interpretation. We characterize the meaning of zero partial correlations in elliptical and elliptical copula models and show that they retain much of the dependence structure from the Gaussian case. Regarding positive dependence, we prove impossibility results to learn certain positive (trans)elliptical graphical models, including that an elliptical distribution that is multivariate totally positive of order two for all dimensions must be essentially Gaussian. We then show how to interpret positive partial correlations as a relaxation, and obtain important properties related to faithfulness and Simpson’s paradox. We illustrate the transelliptical model potential to study tail dependence in S&P500 data, and of positivity to improve regularized inference., DR and PZ were supported from the Spanish Government grants (RYC-2015-18544, RYC2017-22544, PGC2018-101643-B-I00), and Ayudas Fundación BBVA a Equipos de Investigación Cientifica 2017. DR was partially supported by NIH grant R01 CA158113-01 and Europa Excelencia EUR2020-112096.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Density estimates for jump diffusion processes

  • Kohatsu-Higa, Arturo
  • Nualart, Eulàlia
  • Tran, Ngoc Khue
We consider a real-valued diffusion process with a linear jump term driven by a Poisson point process and we assume that the jump amplitudes have a centered density with finite moments. We show upper and lower estimates for the density of the solution in the case that the jump amplitudes follow a Gaussian or Laplacian law. The proof of the lower bound uses a general expression for the density of the solution in terms of the convolution of the density of the continuous part and the jump amplitude density. The upper bound uses an upper tail estimate in terms of the jump amplitude distribution and techniques of the Malliavin calculus in order to bound the density by the tails of the solution. We also extend the lower bounds to the multidimensional case., Eulalia Nualart acknowledges support from the Spanish MINECO grant PGC2018-101643-B-I00 and Ayudas Fundacion BBVA a Equipos de Investigación Científica 2017. Ngoc Khue Tran acknowledges support from the Vietnam Institute for Advanced Study in Mathematics (VIASM ) where a part of this work was done during his visit.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Non-existence results for stochastic wave equations in one dimension

  • Foondun, Mohammud
  • Nualart, Eulàlia
The purpose of this paper is to extend recent results of [2] and [10] for the stochastic heat equation to the stochastic wave equation given by [...] where Ẇ is space-time white noise, σ is a real-valued globally Lipschitz function but b is assumed to be only locally Lipschitz continuous. Three types of domain conditions are studied: D = [0, 1] with homogeneous Dirichlet boundary conditions, D = [0, 2π] with periodic boundary conditions, and D = R. Then, under suitable conditions, the following integrability condition [...] is studied in relation to non-existence of global solutions., EN acknowledges support from the Spanish MINECO grant PGC2018-101643-B-I00 and Ayudas Fundacion BBVA a Equipos de Investigación Científica 2017.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

The neurogenic fate of the hindbrain boundaries relies on Notch3-dependent asymmetric cell divisions

  • Hevia, Covadonga F.
  • Engel-Pizcueta, Carolyn
  • Udina, Frederic
  • Pujades Corbi, Cristina
Elucidating the cellular and molecular mechanisms that regulate the balance between progenitor cell proliferation and neuronal differentiation in the construction of the embryonic brain demands the combination of cell lineage and functional approaches. Here, we generate the comprehensive lineage of hindbrain boundary cells by using a CRISPR-based knockin zebrafish transgenic line that specifically labels the boundaries. We unveil that boundary cells asynchronously engage in neurogenesis undergoing a functional transition from neuroepithelial progenitors to radial glia cells, coinciding with the onset of Notch3 signaling that triggers their asymmetrical cell division. Upon notch3 loss of function, boundary cells lose radial glia properties and symmetrically divide undergoing neuronal differentiation. Finally, we show that the fate of boundary cells is to become neurons, the subtype of which relies on their axial position, suggesting that boundary cells contribute to refine the number and proportion of the distinct neuronal populations., This work was funded by grant PGC2018-095663-B-I00 from Spanish Ministry of Science and Innovation (MICIN), Agencia Estatal de Investigación (AEI), and Fondo Europeo de Desarrollo Regional (FEDER) to C.P. The Department of Medicine and Life Sciences (UPF) is an Unidad de Excelencia María de Maeztu funded by the MICIN and the AEI (DOI: 10.13039/501100011033) Ref: CEX2018-000792-M. C.E.P. is a recipient of a predoctoral FPU fellowship from the Spanish Ministry of Universities. F.U.’s work was supported by PGC2018-101643-B-I00 (MICIN/AEI-FEDER). C.P. is a recipient of ICREA Academia award (Generalitat de Catalunya).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Heterogeneous large datasets integration using Bayesian factor regression

  • Avalos-Pacheco, Alejandra
  • Rossell Ribera, David
  • Savage, Richard S.
Two key challenges in modern statistical applications are the large amount of information recorded per individual, and that such data are often not collected all at once but in batches. These batch effects can be complex, causing distortions in both mean and variance. We propose a novel sparse latent factor regression model to integrate such heterogeneous data. The model provides a tool for data exploration via dimensionality reduction and sparse low-rank covariance estimation while correcting for a range of batch effects. We study the use of several sparse priors (local and non-local) to learn the dimension of the latent factors. We provide a flexible methodology for sparse factor regression which is not limited to data with batch effects. Our model is fitted in a deterministic fashion by means of an EM algorithm for which we derive closed-form updates, contributing a novel scalable algorithm for non-local priors of interest beyond the immediate scope of this paper. We present several examples, with a focus on bioinformatics applications. Our results show an increase in the accuracy of the dimensionality reduction, with non-local priors substantially improving the reconstruction of factor cardinality. The results of our analyses illustrate how failing to properly account for batch effects can result in unreliable inference. Our model provides a novel approach to latent factor regression that balances sparsity with sensitivity in scenarios both with and without batch effects and is highly computationally efficient., Partially funded by the NIH grant R01 CA158113-01, RyC-2015-18544, Ayudas Fundación BBVA a equipos de investigación científica 2017 and Spanish Plan Estatal grant PGC2018-101643-B-I00.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Approximate Laplace approximations for scalable model selection

  • Rossell Ribera, David
  • Abril Pla, Oriol
  • Bhattacharya, Anirban
Includes supplementary materials for the online appendix., We propose the approximate Laplace approximation (ALA) to evaluate integrated likelihoods, a bottleneck in Bayesian model selection. The Laplace approximation (LA) is a popular tool that speeds up such computation and equips strong model selection properties. However, when the sample size is large or one considers many models the cost of the required optimizations becomes impractical. ALA reduces the cost to that of solving a least-squares problem for each model. Further, it enables efficient computation across models such as sharing pre-computed sufficient statistics and certain operations in matrix decompositions. We prove that in generalized (possibly non-linear) models ALA achieves a strong form of model selection consistency for a suitably-defined optimal model, at the same functional rates as exact computation. We consider fixed- and high-dimensional problems, group and hierarchical constraints, and the possibility that all models are misspecified. We also obtain ALA rates for Gaussian regression under non-local priors, an important example where the LA can be costly and does not consistently estimate the integrated likelihood. Our examples include non-linear regression, logistic, Poisson and survival models. We implement the methodology in the R package mombf., Spanish Government grants Europa Excelencia, Grant/Award Number: EUR2020-112096, RYC-2015-18544 and PGC2018-101643-B-I00; NIH, Grant/Award Number: R01 CA158113DMS-01.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Optimal convergence rates for the invariant density estimation of jump-diffusion processes

  • Amorino, Chiara
  • Nualart, Eulàlia
We aim at estimating the invariant density associated to a stochastic differential equation with jumps in low dimension, which is for d = 1 and d = 2. We consider a class of fully non-linear jump diffusion processes whose invariant density belongs to some Hölder space. Firstly, in dimension one, we show that the kernel density estimator achieves the convergence rate 1/T, which is the optimal rate in the absence of jumps. This improves the convergence rate obtained in Amorino and Gloter [J. Stat. Plann. Inference 213 (2021) 106–129], which depends on the Blumenthal-Getoor index for d = 1 and is equal to (logT)/T for d = 2. Secondly, when the jump and diffusion coefficients are constant and the jumps are finite, we show that is not possible to find an estimator with faster rates of estimation. Indeed, we get some lower bounds with the same rates {1/T, (logT)/T} in the mono and bi-dimensional cases, respectively. Finally, we obtain the asymptotic normality of the estimator in the one-dimensional case for the fully non-linear process., CA gratefully acknowledges financial support of ERC Consolidator Grant 815703 “STAMFORD: Statistical Methods for High Dimensional Diffusions”. EN acknowledges support from the Spanish MINECO grant PGC2018-101643-B-I00 and Ayudas Fundación BBVA a Equipos de Investigación Científica 2017.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

General Bayesian loss function selection and the use of improper models

  • Jewson, Jack
  • Rossell Ribera, David
Statisticians often face the choice between using probability models or a paradigm defined by minimising a loss function. Both approaches are useful and, if the loss can be re-cast into a proper probability model, there are many tools to decide which model or loss is more appropriate for the observed data, in the sense of explaining the data's nature. However, when the loss leads to an improper model, there are no principled ways to guide this choice. We address this task by combining the Hyvärinen score, which naturally targets infinitesimal relative probabilities, and general Bayesian updating, which provides a unifying framework for inference on losses and models. Specifically we propose the ℋ-score, a general Bayesian selection criterion and prove that it consistently selects the (possibly improper) model closest to the data-generating truth in Fisher's divergence. We also prove that an associated ℋ-posterior consistently learns optimal hyper-parameters featuring in loss functions, including a challenging tempering parameter in generalised Bayesian inference. As salient examples, we consider robust regression and non-parametric density estimation where popular loss functions define improper models for the data and hence cannot be dealt with using standard model selection tools. These examples illustrate advantages in robustness-efficiency trade-offs and enable Bayesian inference for kernel density estimation, opening a new avenue for Bayesian non-parametrics., Jack Jewson and David Rossell were partially funded by the Ayudas Fundación BBVA a Equipos de Investigación Cientifica 2017 and Proyectos de Investigación Científica en Matemáticas 2021, and Government of Spain's Plan Nacional PGC2018-101643-B-I00 grants. David Rossell was also partially funded by the Europa Excelencia grant EUR2020-112096 and Ramón y Cajal Fellowship RYC-2015-18544.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Instantaneous everywhere-blowup of parabolic SPDEs

  • Foondun, Mohammud
  • Khoshnevisan, Davar
  • Nualart, Eulàlia
Davar Khoshnevisan: Research supported in part by the United States’ National Science Foundation Grants DMS-1855439 and DMS-2245242. Eulalia Nualart: Acknowledges support from the Spanish MINECO Grant PGC2018-101643-B-I00 and Ayudas Fundacion BBVA a Proyectos de Investigación Científica 2021.