Facultad de Ingeniería y Ciencias Aplicadas

Permanent URI for this community

https://repositorio.uandes.cl/handle/uandes/986

Browse

Now showing 1 - 15 of 15

33
(33, 33) Pavez, Lukas; Saavedra, Jose M.; 33
During the last years, we have seen significant advances in the object detection task, mainly due to the outperforming results of convolutional neural networks. In this vein, anchor-based models have achieved the best results. However, these models require prior information about the aspect and scales of target objects, needing more hyperparameters to fit. In addition, using anchors to fit bounding boxes seems far from how our visual system does the same visual task. Instead, our visual system uses the interactions of different scene parts to semantically identify objects, called perceptual grouping. An object detection methodology closer to the natural model is anchor-free detection, where models like FCOS or Centernet have shown competitive results, but these have not yet exploited the concept of perceptual grouping. Therefore, to increase the effectiveness of anchor-free models keeping the inference time low, we propose to add non-local attention (NL modules) modules to boost the feature map of the underlying backbone. NL modules implement the perceptual grouping mechanism, allowing receptive fields to cooperate in visual representation learning. We show that non-local modules combined with an FCOS head (NL-FCOS) are practical and efficient. Thus, we establish state-of-the-art performance in clothing detection and handwritten amount recognition problems.
33
(33, 33) Cedillo-Campos, Miguel Gastón; González-Ramírez, Rosa G.; Camacho-Vallejo, José Fernando; Villalobos, J. Rene; 33
33
33
(33, 33) Hernández Prado, Francisco Javier; Contreras, Nicolás; Zhang, Xihong; Hao, Hong; 33
Interlocking bricks have emerged as a promising alternative to traditional masonry, primarily due to their self-aligning capabilities, which accelerate construction and reduce dependency on skilled labour—advantages particularly valuable for addressing Australia's labour shortages and high costs. However, dry-stacked interlocking brick assemblies inherently exhibit gaps between bricks arising from surface roughness and dimensional variations. Extensive research has shown that these gaps induce stress concentrations and initial non-linear behaviour under static compressive and shear loading while enhancing energy absorption and dissipation under dynamic loading. Despite their known influence on dynamic performance parameters, the impacts of inter-brick gaps on the dynamic response of dry-stacked interlocking bricks during impact loading remain unknown. This study characterises the dynamic response through a combination of experimental testing, analytical simplifications, and stochastic analysis of interlocking brick assemblies. Instrumented impact hammer tests conducted on interlocking assemblies revealed that the inherent gaps cause pounding between adjacent bricks, resulting in high-frequency dynamic responses. An analytical approach was developed to characterise the inter-brick pounding, with results indicating that the timing of pounding events primarily influences the dominant frequency of the assembly. A stochastic method was employed to directly assess the influence of inter-brick gaps on the Fourier spectrum and to establish the boundaries of the analytical analysis. Furthermore, a novel design integrating meta-concrete with interlocking bricks is proposed to exploit these high-frequency responses beneficially. Numerical simulations demonstrated that meta-concrete interlocking brick systems attenuated acceleration responses induced by impact forces by up to 26 %, highlighting the synergy between these technologies for potential applications.
33
(33, 33) Carrasco, Miguel; Vairetti, Carla; López, Julio; Maldonado, Sebastián; 33
Kernel methods are crucial in machine learning due to their ability to model nonlinear relationships in data. Among these, Support Vector Machine (SVM) is widely recognized for its robust performance and appealing optimization properties. In this work, we build upon recent advancements in SVM variants to propose five novel models specifically designed for multiclass learning. In particular, we introduce One-vs-One and One-vs-All versions of the nonparallel hyperplane SVM and improved twin SVM, along with a unified optimization variant (all-together) of the former method for nonlinear multiclass classification. Our empirical evaluation, conducted on 11 datasets and 12 multiclass classifiers, shows the superiority of our methods: four out of the five proposed models rank among the top performers and consistently outperform alternative approaches in terms of balanced accuracy. Additionally, a statistical test was performed, showing significant differences among the classifiers.
33
(33, 33) Rosas, Alexandre; Cisternas, Jaime; Escaff, Daniel; Pinto, Italo'Ivo Lima Dias; Lindenberg, Katja; 33
It is well established that ensembles of globally coupled stochastic oscillators may exhibit a nonequilibrium phase transition to synchronization in the thermodynamic limit (infinite number of elements). In fact, since the early work of Kuramoto, mean-field theory has been used to analyze this transition. In contrast, work that directly deals with finite arrays is relatively scarce in the context of synchronization. And yet it is worth noting that finite-number effects should be seriously taken into account since, in general, the limits N?? (where N is the number of units) and t?? (where t is time) do not commute. Mean-field theory implements the particular choice first N?? and then t??. Here we analyze an ensemble of three-state coupled stochastic units, which has been widely studied in the thermodynamic limit. We formally address the finite-N problem by deducing a Fokker-Planck equation that describes the system. We compute the steady-state solution of this Fokker-Planck equation (that is, finite N but t??). We use this steady state to analyze the synchronic properties of the system in the framework of the different order parameters that have been proposed in the literature to study nonequilibrium transitions.
33
(33, 33) Rohe, K.; Cisternas, J.; 33
Reaction-diffusion systems are used in biology, chemistry, and physics to model the interaction of spatially distributed species. Particularly of interest is the spatial replacement of one equilibrium state by another, depicted as traveling waves or fronts. Their profiles and traveling velocity depend on the nonlinearities in the reaction term and on spatial diffusion. If the reaction occurs at regularly spaced points, the velocities also depend on lattice structures and the orientation of the traveling front. Interestingly, there is a wide region of parameters where the speeds become zero and the fronts do not propagate. In this paper, we focus on systems with three stable coexisting equilibrium states that are described by the butterfly bifurcation and study to what extent the three possible 1D traveling fronts suffer from propagation failure. We demonstrate that discreteness of space affects the three fronts differently. Regions of propagation failure add a new layer of complexity to the butterfly diagram. The analysis is extended to planar fronts traveling through different orientations in regular 2D lattices. Both propagation failure and the existence of preferred orientations play a role in the transient and long-time evolution of 2D patterns.
33
(33, 33) González-Ramírez, Rosa G.; Leung, Janny; Otto, Alena; Pesch, Erwin; 33
33
33
(33, 33) España, Francisco; Scott, Felipe; 33
33
33
(33, 33) Cisternas, Jaime; Mellado, Paula; Urbina, Felipe; Portilla, Cristóbal; Carrasco, Miguel; Concha, Andrés; 33
In classical mechanics, solutions can be classified according to their stability. Each of them is part of the possible trajectories of the system. However, the signatures of unstable solutions are hard to observe in an experiment, and most of the times if the experimental realization is adiabatic, they are considered just a nuisance. Here we use a small number of XY magnetic dipoles subject to an external magnetic field for studying the origin of their collective magnetic response. Using bifurcation theory we have found all the possible solutions being stable or unstable, and explored how those solutions are naturally connected by points where the symmetries of the system are lost or restored. Unstable solutions that reveal the symmetries of the system are found to be the culprit that shape hysteresis loops in this system. The complexity of the solutions for the nonlinear dynamics is analyzed using the concept of boundary basin entropy, finding that the damping timescale is critical for the emergence of fractal structures in the basins of attraction. Furthermore, we numerically found domain wall solutions that are the smallest possible realizations of transverse walls and vortex walls in magnetism. We experimentally confirmed their existence and stability showing that our system is a suitable platform to study domain wall dynamics at the macroscale.
33
(Springer Nature Switzerland AG, 33) Echeverría, Antonia; Del Solar, Paulina; Fernández, Rodrigo; 33
The population growth of older people is one of the main challenges in how countries respond to the specific demands of aging, where the main problem is the increase and progression of older people’s dependence. The use of public transport has been called in the literature as the “gateway” to dependency. Transportation is an essential element to be studied in countries trying to implement age-friendly cities that favour healthy aging and old-age functionality. The purpose of this chapter is to describe the relationship between the functionality of older people and their opportunities to participate in significant environments, analysing the impact on their day-to-day activities in the community using public transport as a means to shape friendly cities. Through a literature review, we will explore the implications for aging, functionality, and transportation policies, considering Chile as a case study. The understanding of the issue from an interdisciplinary approach, considering the academic and scientific evidence and the Latin American country with the highest rate of population aging as a case study, makes available an opportunity for countries that are about to experience a similar situation to learn based on experience how to tackle this problem. This chapter raises awareness among professionals, academics, and researchers about the elements related to transportation, its implications, and repercussions on the older population.
33
(33, 33) Cisternas, J.; Navarro, M.; Duarte, S.; Concha, A.; 33
Macroscopic magnets can easily be manipulated and positioned so that interactions between themselves and with external fields induce interesting dynamics and equilibrium configurations. In this work, we use rotating magnets positioned in a line or at the vertices of a regular polygon. The rotation planes of the magnets can be modified at will. The rich structure of stable and unstable configurations is dictated by symmetry and the side of the polygon. We show that both symmetric solutions and their symmetry-breaking bifurcations can be explained with group theory. Our results suggest that the predicted magnetic textures should emerge at any length scale as long as the interaction is polar, and the system is endowed with the same symmetries.
33
(33, 33) Liu, Haoyuan; Diambra, Andrea; Abell, José Antonio; Pisanò, Federico; 33
This work presents a critical state plasticity model for predicting the response of sands to cyclic loading. The well-known bounding surface SANISAND framework by Dafalias and Manzari is enhanced with a memory surface to capture micromechanical, fabric-related processes directly affecting cyclic sand behavior. The resulting model, SANISAND-MS, was recently proposed by Liu et al. and successfully applied to the simulation of drained sand ratcheting under thousands of loading cycles. Herein, novel ingredients are embedded into Liu et al.'s formulation to better capture the effects of fabric evolution history on sand stiffness and dilatancy. The new features enable remarkable accuracy in simulating undrained pore pressure buildup and cyclic mobility behavior in medium-dense to dense sand. The performance of the upgraded SANISAND-MS is validated against experimental test results from the literature-including undrained cyclic triaxial tests at varying cyclic loading conditions and precyclic consolidation histories. The proposed modeling platform will positively impact the study of relevant cyclic and dynamic problems, for instance, in the fields of earthquake and offshore geotechnics.
33
(33, 33) Cedillo-Campos, Miguel Gaston; González-Ramírez, Rosa G.; Mejía-Argueta, Christopher; González-Feliu, Jesús; 33
33
33
(33, 33) Morales, Javier; Murrugarra-Llerena, Nils; Saavedra, Jose M.; 33
Sketch-based understanding is a critical component of human cognitive learning and is a primitive communication means between humans. This topic has recently attracted the interest of the computer vision community as sketching represents a powerful tool to express static objects and dynamic scenes. Unfortunately, despite its broad application domains, the current sketch-based models strongly rely on labels for supervised training, ignoring knowledge from unlabeled data, thus limiting the underlying generalization and the applicability. Therefore, we present a study about the use of unlabeled data to improve a sketch-based model. To this end, we evaluate variations of VAE and semi-supervised VAE, and present an extension of BYOL to deal with sketches. Our results show the superiority of sketch-BYOL, which outperforms other self-supervised approaches increasing the retrieval performance for known and unknown categories. Furthermore, we show how other tasks can benefit from our proposal.
33
(33, 33) Verma, Anant Kumar; Zuloaga, Rodrigo; Levis Calluil, Héctor Matías; Martinez-Gomez, Manuel; Burgos-Mellado, Claudio; Torres, Miguel A.; 33
33

Browse

Browsing Facultad de Ingeniería y Ciencias Aplicadas by Subject "33"

Results Per Page

Sort Options