Showing 1–10 of 10 results for author: Munoz, I

Optimizing sDTW for AMD GPUs

Authors: Daniel Latta-Lin, Sofia Isadora Padilla Munoz

Abstract: Subsequence Dynamic Time War** (sDTW) is the metric of choice when performing many sequence matching and alignment tasks. While sDTW is flexible and accurate, it is neither simple nor fast to compute; significant research effort has been spent devising parallel implementations on the GPU that leverage efficient memory access and computation patterns, as well as features offered by specific vendo… ▽ More Subsequence Dynamic Time War** (sDTW) is the metric of choice when performing many sequence matching and alignment tasks. While sDTW is flexible and accurate, it is neither simple nor fast to compute; significant research effort has been spent devising parallel implementations on the GPU that leverage efficient memory access and computation patterns, as well as features offered by specific vendors and architectures (notably NVIDIA's). We present an implementation of sDTW on AMD hardware using HIP and ROCm. Our implementation employs well-known parallel patterns, as well as lower-level features offered by ROCm. We use shuffling for intra-wavefront communication and shared memory to transfer data between consecutive wavefronts. By constraining the input data to batches of 512 queries of length 2,000, we optimized for peak performance the width of reference elements operated on by a single thread. △ Less

Submitted 11 March, 2024; originally announced March 2024.

Platformization of Inequality: Gender and Race in Digital Labor Platforms

Authors: Isabel Munoz, Pyeonghwa Kim, Clea O'Neil, Michael Dunn, Steve Sawyer

Abstract: We contribute empirical and conceptual insights regarding the roles of digital labor platforms in online freelancing, focusing attention to social identities such as gender, race, ethnicity, and occupation. Findings highlight how digital labor platforms reinforce and exacerbate identity-based stereotypes, bias and expectations in online freelance work. We focus on online freelancing as this form o… ▽ More We contribute empirical and conceptual insights regarding the roles of digital labor platforms in online freelancing, focusing attention to social identities such as gender, race, ethnicity, and occupation. Findings highlight how digital labor platforms reinforce and exacerbate identity-based stereotypes, bias and expectations in online freelance work. We focus on online freelancing as this form of working arrangement is becoming more prevalent. Online freelancing also relies on the market-making power of digital platforms to create an online labor market. Many see this as one likely future of work with less bias. Others worry that labor platforms' market power allows them to embed known biases into new working arrangements: a platformization of inequality. Drawing on data from 108 online freelancers, we discuss six findings: 1) female freelance work is undervalued; 2) gendered occupational expectations; 3) gendered treatment; 4) shared expectations of differential values; 5) racial stereotypes and expectations; and 6) race and ethnicity as an asset. We discuss the role of design in the platformization and visibility of social identity dimensions and the implications of the reinforced identity perceptions and marginalization in digital labor platforms. △ Less

Submitted 28 September, 2023; originally announced September 2023.

Comments: PAPER WAS ACCEPTED AT CSCW ON SEPTEMBER 2023, 21 PAGES, 2 FIGURES

Some qualitative properties of solutions to a reaction-diffusion equation with weighted strong reaction

Authors: Razvan Gabriel Iagar, Ana Isabel Muñoz, Ariel Sánchez

Abstract: We study the existence and qualitative properties of solutions to the Cauchy problem associated to the quasilinear reaction-diffusion equation $$ \partial_tu=Δu^m+(1+|x|)^σu^p, $$ posed for $(x,t)\in\real^N\times(0,\infty)$, where $m>1$, $p\in(0,1)$ and $σ>0$. Initial data are taken to be bounded, non-negative and compactly supported. In the range when $m+p\geq2$, we prove \emph{local existence of… ▽ More We study the existence and qualitative properties of solutions to the Cauchy problem associated to the quasilinear reaction-diffusion equation $$ \partial_tu=Δu^m+(1+|x|)^σu^p, $$ posed for $(x,t)\in\real^N\times(0,\infty)$, where $m>1$, $p\in(0,1)$ and $σ>0$. Initial data are taken to be bounded, non-negative and compactly supported. In the range when $m+p\geq2$, we prove \emph{local existence of solutions} together with a \emph{finite speed of propagation} of their supports for compactly supported initial conditions. We also show in this case that, for a given compactly supported initial condition, there exist \emph{infinitely many solutions} to the Cauchy problem, by prescribing the evolution of their interface. In the complementary range $m+p<2$, we establish new \emph{Aronson-Bénilan estimates} satisfied by solutions to the Cauchy problem, which are of independent interest as a priori bounds for the solutions. We apply these estimates to establish \emph{infinite speed of propagation} of the supports of solutions if $m+p<2$, that is, $u(x,t)>0$ for any $x\in\real^N$, $t>0$, even in the case when the initial condition $u_0$ was compactly supported. △ Less

Submitted 13 June, 2023; v1 submitted 1 April, 2023; originally announced April 2023.

Extinction and non-extinction profiles for the sub-critical fast diffusion equation with weighted source

Authors: Razvan Gabriel Iagar, Ana Isabel Muñoz, Ariel Sánchez

Abstract: We establish both extinction and non-extinction self-similar profiles for the following fast diffusion equation with a weighted source term $$ \partial_tu=Δu^m+|x|^σu^p, $$ posed for $(x,t)\in\real^N\times(0,\infty)$, $N\geq3$, in the sub-critical range of the fast diffusion equation $0<m<m_c=(N-2)/N$. We consider $σ>0$ and $\max\{p_c(σ),1\}<p<p_L(σ)$, where… ▽ More We establish both extinction and non-extinction self-similar profiles for the following fast diffusion equation with a weighted source term $$ \partial_tu=Δu^m+|x|^σu^p, $$ posed for $(x,t)\in\real^N\times(0,\infty)$, $N\geq3$, in the sub-critical range of the fast diffusion equation $0<m<m_c=(N-2)/N$. We consider $σ>0$ and $\max\{p_c(σ),1\}<p<p_L(σ)$, where $$ p_c(σ)=\frac{m(N+σ)}{N-2}, \qquad p_L(σ)=1+\frac{σ(1-m)}{2}. $$ We show that, on the one hand, positive self-similar solutions at any time $t>0$, in the form $$ u(x,t)=t^αf(|x|t^β), \qquad f(ξ)\sim Cξ^{-(N-2)/m}, \qquad α>0, \ β>0 $$ exist, provided $0<m<m_s=(N-2)/(N+2)$ and $p_s(σ)=m(N+2σ+2)/(N-2)<p<p_L(σ)$. On the other hand, we prove that there exists $p_0(σ)\in(p_c(σ),p_s(σ))$ such that self-similar solutions presenting finite time extinction are established both for $p\in(p_0(σ),p_s(σ))$ and for $p\in(p_s(σ),p_L(σ))$, but with profiles $f(ξ)$ having different spatially decreasing tails as $|x|\to\infty$. We also prove non-existence of self-similar solutions in complementary ranges of exponents to the ones described above or if $m\geq m_c$. △ Less

Submitted 19 February, 2023; originally announced February 2023.

MSC Class: 35B33; 35B36; 35C06; 35K57; 35K59

Bulk renormalization and the AdS/CFT correspondence

Authors: Máximo Bañados, Ernesto Bianchi, Iván Muñoz, Kostas Skenderis

Abstract: We develop a systematic renormalization procedure for QFT in anti-de Sitter spacetime. UV infinities are regulated using a geodesic point-splitting method, which respects AdS isometries, while IR infinities are regulated by cutting off the radial direction (as in holographic renormalization). The renormalized theory is defined by introducing $Z$ factors for all parameters in the Lagrangian and the… ▽ More We develop a systematic renormalization procedure for QFT in anti-de Sitter spacetime. UV infinities are regulated using a geodesic point-splitting method, which respects AdS isometries, while IR infinities are regulated by cutting off the radial direction (as in holographic renormalization). The renormalized theory is defined by introducing $Z$ factors for all parameters in the Lagrangian and the boundary conditions of bulk fields (sources of dual operators), and a boundary counterterm action, $S_{\rm ct}$, such that the limit of removing the UV and IR regulators exists. The results are in general scheme dependent (mirroring the analogous result in flat space) and require renormalization conditions. These may be provided by the dual CFT (or by string theory in AdS). Our analysis amounts also to a first principles derivation of the Feynman rules regarding Witten diagrams. The presence and treatment of IR divergences is essential for correctly accounting for anomalous dimensions of dual operators. We apply the method to scalar $Φ^4$ theory and obtain the renormalized 2-point function of the dual operator to 2-loops, and the renormalized 4-point function to 1-loop order, for operators of any dimension $Δ$ and bulk spacetime dimension up to $d+1=7$. △ Less

Submitted 22 January, 2023; v1 submitted 24 August, 2022; originally announced August 2022.

Comments: 7 pages, 1 figure; v2: Comments and references added. Published version

Spectral Variational Multi-Scale method for parabolic problems. Application to 1D transient advection-diffusion equations

Authors: Tomás Chacón Rebollo, Soledad Fernández-García, David Moreno-Lopez, Isabel Sánchez Muñoz

Abstract: In this work, we introduce a Variational Multi-Scale (VMS) method for the numerical approximation of parabolic problems, where sub-grid scales are approximated from the eigenpairs of associated elliptic operator. The abstract method is particularized to the one-dimensional advection-diffusion equations, for which the sub-grid components are exactly calculated in terms of a spectral expansion when… ▽ More In this work, we introduce a Variational Multi-Scale (VMS) method for the numerical approximation of parabolic problems, where sub-grid scales are approximated from the eigenpairs of associated elliptic operator. The abstract method is particularized to the one-dimensional advection-diffusion equations, for which the sub-grid components are exactly calculated in terms of a spectral expansion when the advection velocity is approximated by piecewise constant velocities on the grid elements. We prove error estimates that in particular imply that when Lagrange finite element discretisations in space are used, the spectral VMS method coincides with the exact solution of the implicit Euler semi-discretisation of the advection-diffusion problem at the Lagrange interpolation nodes. We also build a feasible method to solve the evolutive advection-diffusion problems by means of an offline/online strategy with reduced computational complexity. We perform some numerical tests in good agreement with the theoretical expectations, that show an improved accuracy with respect to several stabilised methods. △ Less

Submitted 21 July, 2022; originally announced July 2022.

Self-similar solutions preventing finite time blow-up for reaction-diffusion equations with singular potential

Authors: Razvan Gabriel Iagar, Ana Isabel Muñoz, Ariel Sánchez

Abstract: We prove existence and uniqueness of a global in time self-similar solution growing up as $t\to\infty$ for the following reaction-diffusion equation with a singular potential $$ u_t=Δu^m+|x|^σu^p, $$ posed in dimension $N\geq2$, with $m>1$, $σ\in(-2,0)$ and $1<p<1-σ(m-1)/2$. For the special case of dimension $N=1$, the same holds true for $σ\in(-1,0)$ and similar ranges for $m$ and $p$. The existe… ▽ More We prove existence and uniqueness of a global in time self-similar solution growing up as $t\to\infty$ for the following reaction-diffusion equation with a singular potential $$ u_t=Δu^m+|x|^σu^p, $$ posed in dimension $N\geq2$, with $m>1$, $σ\in(-2,0)$ and $1<p<1-σ(m-1)/2$. For the special case of dimension $N=1$, the same holds true for $σ\in(-1,0)$ and similar ranges for $m$ and $p$. The existence of this global solution prevents finite time blow-up even with $m>1$ and $p>1$, showing an interesting effect induced by the singular potential $|x|^σ$. This result is also applied to reaction-diffusion equations with general potentials $V(x)$ to prevent finite time blow-up via comparison. △ Less

Submitted 8 November, 2021; originally announced November 2021.

Self-similar blow-up patterns for a reaction-diffusion equation with weighted reaction in general dimension

Authors: Razvan Gabriel Iagar, Ana I. Muñoz, Ariel Sánchez

Abstract: We classify the finite time blow-up profiles for the following reaction-diffusion equation with unbounded weight: $$ \partial_tu=Δu^m+|x|^σu^p, $$ posed in any space dimension $x\in\mathbf{R}^N$, $t\geq0$ and with exponents $m>1$, $p\in(0,1)$ and $σ>2(1-p)/(m-1)$. We prove that blow-up profiles in backward self-similar form exist for the indicated range of parameters, showing thus that the unbound… ▽ More We classify the finite time blow-up profiles for the following reaction-diffusion equation with unbounded weight: $$ \partial_tu=Δu^m+|x|^σu^p, $$ posed in any space dimension $x\in\mathbf{R}^N$, $t\geq0$ and with exponents $m>1$, $p\in(0,1)$ and $σ>2(1-p)/(m-1)$. We prove that blow-up profiles in backward self-similar form exist for the indicated range of parameters, showing thus that the unbounded weight has a strong influence on the dynamics of the equation, merging with the nonlinear reaction in order to produce finite time blow-up. We also prove that all the blow-up profiles are \emph{compactly supported} and might present two different types of interface behavior and three different possible \emph{good behaviors} near the origin, with direct influence on the blow-up behavior of the solutions. We classify all these profiles with respect to these different local behaviors depending on the magnitude of $σ$. This paper generalizes in dimension $N>1$ previous results by the authors in dimension $N=1$ and also includes some finer classification of the profiles for $σ$ large that is new even in dimension $N=1$. △ Less

Submitted 20 August, 2021; originally announced August 2021.

MSC Class: 35A24; 35B44; 35C06; 35K10; 35K65

Diseño y desarrollo de aplicación móvil para la clasificación de flora nativa chilena utilizando redes neuronales convolucionales

Authors: Ignacio Muñoz, Alfredo Bolt

Abstract: Introduction: Mobile apps, through artificial vision, are capable of recognizing vegetable species in real time. However, the existing species recognition apps do not take in consideration the wide variety of endemic and native (Chilean) species, which leads to wrong species predictions. This study introduces the development of a chilean species dataset and an optimized classification model implem… ▽ More Introduction: Mobile apps, through artificial vision, are capable of recognizing vegetable species in real time. However, the existing species recognition apps do not take in consideration the wide variety of endemic and native (Chilean) species, which leads to wrong species predictions. This study introduces the development of a chilean species dataset and an optimized classification model implemented to a mobile app. Method: the data set was built by putting together pictures of several species captured on the field and by selecting some pictures available from other datasets available online. Convolutional neural networks were used in order to develop the images prediction models. The networks were trained by performing a sensitivity analysis, validating with k-fold cross validation and performing tests with different hyper-parameters, optimizers, convolutional layers, and learning rates in order to identify and choose the best models and then put them together in one classification model. Results: The final data set was compounded by 46 species, including native species, endemic and exotic from Chile, with 6120 training pictures and 655 testing pictures. The best models were implemented on a mobile app, obtaining a 95% correct prediction rate with respect to the set of tests. Conclusion: The app developed in this study is capable of classifying species with a high level of accuracy, depending on the state of the art of the artificial vision and it can also show relevant information related to the classified species. △ Less

Submitted 10 December, 2021; v1 submitted 11 June, 2021; originally announced June 2021.

Comments: in Spanish

Ultracool Dwarfs in deep extragalactic surveys using the Virtual Observatory: ALHAMBRA and COSMOS

Authors: E. Solano, M. C. Gálvez-Ortiz, E. L. Martín, I. M. Gómez Muñoz, C. Rodrigo, A. J. Burgasser, N. Lodieu, V. J. S. Béjar, N. Huélamo, M. Morales-Calderón, H. Bouy

Abstract: Ultracool dwarfs encompass a wide variety of compact stellar-like objects with spectra classified as late-M, L, T and Y. Most of them have been discovered using wide-field imaging surveys. The Virtual Observatory has proven to be of great utility to efficiently exploit these astronomical resources. We aim to validate a Virtual Observatory methodology designed to discover and characterize ultracool… ▽ More Ultracool dwarfs encompass a wide variety of compact stellar-like objects with spectra classified as late-M, L, T and Y. Most of them have been discovered using wide-field imaging surveys. The Virtual Observatory has proven to be of great utility to efficiently exploit these astronomical resources. We aim to validate a Virtual Observatory methodology designed to discover and characterize ultracool dwarfs in deep extragalactic surveys like ALHAMBRA and COSMOS. Three complementary searches based on parallaxes, proper motions and colours, respectively were carried out. A total of 897 candidate ultracool dwarfs were found, with only 16 previously reported in SIMBAD. Most of the new UCDs reported here are likely late-M and L dwarfs because of the limitations imposed by the utilization of optical ($Gaia$ DR2 and r-band) data. We complement ALHAMBRA and COSMOS photometry with other catalogues in the optical and infrared using VOSA, a Virtual Observatory tool that estimates effective temperatures from the spectral energy distribution fitting to collections of theoretical models. The agreement between the number of UCDs found in the COSMOS field and theoretical estimations together with the low false negative rate (known UCDs not discovered in our search) validates the methodology proposed in this work, which will be used in the forthcoming wide and deep surveys provided by the Euclid space mission. Simulations of Euclid number counts for UCDs detectable in different photometric passbands are presented for a wide survey area of 15,000 square degrees, and the limitations of applicability of Euclid data to detect UCDs using the methods employed in this paper are discussed. △ Less

Submitted 30 October, 2020; originally announced October 2020.