University of Mendoza

Resumen es: La enfermedad de Kawasaki es una enfermedad febril aguda infantil. La morbimortalidad se relaciona con la existencia de aneurismas coronarios. Se p...

Key challenges in running a retail business include how to select products to present to consumers (the assortment problem), and how to price products (the pricing problem) to maximize revenue or profit. Instead of considering these problems in isolation, we propose a joint approach to assortment-pricing based on contextual bandits. Our model is doubly high-dimensional, in that both context vectors and actions are allowed to take values in high-dimensional spaces. In order to circumvent the curse of dimensionality, we propose a simple yet flexible model that captures the interactions between covariates and actions via a (near) low-rank representation matrix. The resulting class of models is reasonably expressive while remaining interpretable through latent factors, and includes various structured linear bandit and pricing models as particular cases. We propose a computationally tractable procedure that combines an exploration/exploitation protocol with an efficient low-rank matrix estimator, and we prove bounds on its regret. Simulation results show that this method has lower regret than state-of-the-art methods applied to various standard bandit and pricing models. Real-world case studies on the assortment-pricing problem, from an industry-leading instant noodles company to an emerging beauty start-up, underscore the gains achievable using our method. In each case, we show at least three-fold gains in revenue or profit by our bandit method, as well as the interpretability of the latent factor models that are learned.

This study investigates the mechanical behavior of BCC HfNbTaTiZr high-entropy alloy (HEA) nanoparticles (NPs) subjected to compression by a flat indenter using molecular dynamics simulations. NPs with random atomic distribution and diameters from 10 to 50 nm were considered. For comparison purposes, a 20 nm NP with chemical short-range order (CSRO) was also explored. The mechanical response revealed an increasing trend of the Young's modulus with respect to the NP size. However, maximum stress, yield stress, and flow stress showed negligible variations. Furthermore, the NP with CSRO showed enhanced mechanical properties, attributed to SRO cluster formation. Analysis of plastic activity revealed surface-dominated dislocation emission and absorption, with CSRO NP exhibiting an increase in dislocation density as the strain increased. Structural analysis elucidated persistent twin formation in all NPs, with a remarkable increase in the HCP population observed in the CSRO NP. These findings further improve our understanding of the mechanical behavior of HEA NPs, contributing to the design and development of advanced materials.

Coronal mass ejections (CMEs) are a major driver of space weather. To assess CME geoeffectiveness, among other scientific goals, it is necessary to reliably identify and characterize their morphology and kinematics in coronagraph images. Current methods of CME identification are either subjected to human biases or perform a poor identification due to deficiencies in the automatic detection. In this approach, we have trained the deep convolutional neural model Mask R-CNN to automatically segment the outer envelope of one or multiple CMEs present in a single difference coronagraph image. The empirical training dataset is composed of 1.13× 10^5 synthetic coronagraph images with known pixel-level CME segmentation masks. It is obtained by combining quiet (no CME) coronagraph observations, with synthetic white-light CMEs produced using the Graduated Cylindrical Shell geometric model and ray-tracing technique. To filter the different instances found by Mask R-CNN, we use the temporal consistency of mask properties such as the intersection over union ( IoU ). We found that our model-based trained Mask R-CNN infers segmentation masks that are smooth and topologically connected (without holes or isolated patches). While the inferred masks are not representative of the detailed outer envelope of complex CMEs, the neural model can better differentiate a CME from other radially moving background/foreground features, segment multiple simultaneous CMEs that are close to each other, and work with images from different instruments. This is accomplished without relying on kinematic information, i.e. only the included in the single input difference image. We obtain a median IoU=0.98 for 1.6× 10^4 synthetic validation images, and IoU=0.77 when compared with two independent manual segmentations of 115 observations acquired by the COR2-A, COR2-B, and LASCO C2 coronagraphs. The methodology presented in this work can be used with other CME models to produce more realistic synthetic brightness images while preserving desired morphological features, and obtain more robust and/or tailored segmentations.

Magnetic domain wall (DW) motion in nanowires (NWs) is highly sensitive to structural disorder, yet most simulation studies assume ideal crystalline geometries or impose defect effects phenomenologically. NWs inevitably include surface relaxation and contain lattice defects, and may also undergo mechanical deformation during fabrication or operation. In this work, we employ spin-lattice dynamics (SLD) simulations to investigate DW motion in iron NWs and to show that voids and plastic deformation can control DW mobility, enabling an atomistic description of the coupled evolution of spins and lattice degrees of freedom. We first analyze DW motion in crystalline Fe NWs and demonstrate that SLD reproduces analytical predictions for DW width and velocity across a range of magnetic fields, anisotropy values, and damping parameters, with deviations of less than 20%. We then extend the study to NW containing a nanoscale void and plastically deformed structures generated under compression. These large-scale simulations employ about 425,000 atoms/spins. The atomistic features, including voids, dislocations, and grain boundaries, lead to non-monotonic DW motion in a NW and fluctuations during propagation, which are governed by a heterogeneous magnetic energy landscape that varies with lattice disorder. These effects are difficult to capture by conventional micromagnetic approaches. This study highlights the critical role of atomic-scale defects in controlling DW mobility and demonstrates that spin-lattice dynamics is especially well suited for accurately describing magnetization dynamics in nanostructures. Our atomic results establish that defect-generated lattice disorder creates a heterogeneous magnetic-energy landscape that controls DW motion and is difficult to represent predictively within standard continuum models.