Search

Social media images have proven to be a valuable source of information for understanding human interactions with important subjects such as cultural heritage, biodiversity, and nature, among others. The task of grouping such images into a number of semantically meaningful clusters without labels is challenging due to the high diversity and complex nature of the visual content in addition to their large volume. On the other hand, recent advances in Large Visual Models (LVMs), Large Language Models (LLMs), and Large Visual Language Models (LVLMs) provide an important opportunity to explore new productive and scalable solutions. This work proposes, analyzes, and compares various approaches based on one or more state-of-the-art LVM, LLM, and LVLM, for mapping social media images into a number of predefined classes. As a case study, we consider the problem of understanding the interactions between humans and nature, also known as Nature's Contribution to People or Cultural Ecosystem Services (CES). Our experiments show that the highest-performing approaches, with accuracy above 95%, still require the creation of a small labeled dataset. These include the fine-tuned LVM DINOv2 and the LVLM LLaVA-1.5 combined with a fine-tuned LLM. The top fully unsupervised approaches, achieving accuracy above 84%, are the LVLMs, specifically the proprietary GPT-4 model and the public LLaVA-1.5 model. Additionally, the LVM DINOv2, when applied in a 10-shot learning setup, delivered competitive results with an accuracy of 83.99%, closely matching the performance of the LVLM LLaVA-1.5., Comment: 17 pages, 9 figures

Novel brain biomarkers of obesity were sought by studying statistical measurements on fractional anisotropy (FA) images of different white matter (WM) tracts from subjects with specific demographic characteristics. Tract measurements were chosen that showed differences between two groups (normal weigh and overweight/obese) and that were correlated with their BMI. From these measurements, a simple and novel process was applied to select those that would allow the creation of models to quantify and classify the state of obesity of individuals. The biomarkers were created from the tract measurements used in the models. Some positive correlations were found between WM integrity and BMI, mainly in tracts involved in motor functions. From this result, neuroplasticity in motor tracts associated with obesity was hypothesized. Two models were built to quantify and classify obesity status, whose regression coefficients formed the novel proposed obesity-associated brain biomarkers. A process for the selection of tract measurements was proposed, such that models were built to determine the obesity status of subjects individually. From these models, novel brain biomarkers associated with obesity were created. These allow the generation of new knowledge and are intended to be a future tool in the clinical environment for the prevention and treatment of brain changes associated with obesity.

In this paper, we explore the application of Machine Learning techniques, specifically Support Vector Machines (SVM), to unveil the chaotic and regular nature of trajectories in Hamiltonian systems using Lagrangian descriptors. Traditional chaos indicators, while effective, are computationally expensive and require an exhaustive study of the parameter space to establish the classification thresholds. By using SVMs trained on a dataset obtained from the analysis of the dynamics of the double pendulum Hamiltonian system, we aim at reducing the complexity of this process. Our trained SVM models demonstrate high accuracy when it comes to classifying trajectories in diverse Hamiltonian systems, such as for example in the four-well Hamiltonian, the H\'enon-Heiles system and the Chirikov Standard Map. The results indicate that SVMs, when combined with Lagrangian descriptors, offer a robust and efficient method for chaos classification across different dynamical systems. Our approach not only simplifies the classification process but also is highlighting the potential of Machine Learning algorithms in the study of nonlinear dynamics and chaos.

In this paper we apply the method of Lagrangian descriptors as an indicator to study the chaotic and regular behavior of trajectories in the phase space of the classical double pendulum system. In order to successfully quantify the degree of chaos with this tool, we first derive Hamilton's equations of motion for the problem in non-dimensional form, showing that they can be written compactly using matrix algebra. Once the dynamical equations are obtained, we carry out a parametric study in terms of the system's total energy and the other model parameters (lengths and masses of the pendulums, and gravity), to determine the extent of the chaotic and regular regions in the phase space. Our numerical results show that for a given mass ratio, the maximum chaotic fraction of phase space trajectories is attained when the pendulums have equal lengths. Moreover, we give a characterization of the growth and decay of chaos in the system in terms of the model parameters, and explore the hypothesis that the chaotic fraction follows an exponential law over different energy regimes., Comment: 17 pages, 7 figures

The interactions of charged particles moving paraxially in multi-walled carbon nanotubes (MWCNTs) may excite electromagnetic modes. This wake effect has recently been proposed as a potential novel method of short-wavelength high-gradient particle acceleration. In this work, the excitation of wakefields in double-walled carbon nanotubes (DWCNTs) is studied by means of the linearized hydrodynamic theory. General expressions have been derived for the excited longitudinal and transverse wakefields and related to the resonant wavenumbers which can be obtained from the dispersion relation. In the absence of friction, the stopping power of the wakefield driver, modelled here as a charged macroparticle, can be written solely as a function of these resonant wavenumbers. The dependencies of the wakefields on the radii of the DWCNT and the driving velocity have been studied. DWCNTs with inter-wall distances much smaller than the internal radius may be a potential option to obtain higher wakefields for particle acceleration compared to single-walled carbon nanotubes (SWCNTs)., Comment: arXiv admin note: text overlap with arXiv:2308.08399

Quantification of microstructures is crucial for understanding processing-structure and structure-property relationships in polycrystalline materials. Delineating grain boundaries in bright-field transmission electron micrographs, however, is challenging due to complex diffraction contrast in images. Conventional edge detection algorithms are inadequate; instead, manual tracing is usually required. This study demonstrates the first successful machine-learning approach for grain-boundary detection in bright-field transmission electron micrographs. The proposed methodology uses a U-Net convolutional neural network trained on carefully constructed data from bright-field images and hand-tracings available from prior studies, combined with targeted post-processing algorithms to preserve fine features of interest. The image processing pipeline accurately estimates grain-boundary positions, avoiding segmentation in regions with intragrain contrast and identifying low-contrast boundaries. Our approach is validated by directly comparing microstructural markers (i.e., grain centroids) identified in U-Net predictions with those identified in hand tracings; furthermore, the grain size distributions obtained from the two techniques show notable overlap when compared using t-, Kolmogorov-Smirnov, and Cramer-von Mises tests. The technique is then successfully applied to interpret new aluminum film microstructures having different image characteristics from the training data, and preliminary results from Pt and Pd microstructures are presented, highlighting the versatility of our approach for grain-boundary identification in bright-field micrographs.

Childhood overweight and obesity is one of the main health problems in the world since it is related to the early appearance of different diseases, in addition to being a risk factor for later developing obesity in adulthood with its health and economic consequences. Visceral abdominal tissue (VAT) is strongly related to the development of metabolic and cardiovascular diseases compared to abdominal subcutaneous adipose tissue (ASAT). Therefore, precise and automatic VAT and ASAT quantification methods would allow better diagnosis, monitoring and prevention of diseases caused by obesity at any stage of life. Currently, magnetic resonance imaging is the standard for fat quantification, with Dixon sequences being the most useful. Different semiautomatic and automatic ASAT and VAT quantification methodologies have been proposed. In particular, the semi-automated quantification methodology used commercially through the cloud-based service AMRA R Researcher stands out due to its extensive validation in different studies. In the present work, a database made up of Dixon MRI sequences, obtained from children between 7 and 9 years of age, was studied. Applying a preprocessing to obtain what we call total intensity maps, a convolutional neural network (CNN) was proposed for the automatic quantification of ASAT and VAT. The quantifications obtained from the proposed methodology were compared with quantifications previously made through AMRA R Researcher. For the comparison, correlation analysis, Bland-Altman graphs and non-parametric statistical tests were used. The results indicated a high correlation and similar precisions between the quantifications of this work and those of AMRA R Researcher. The final objective is that the proposed methodology can serve as an accessible and free tool for the diagnosis, monitoring and prevention of diseases related to childhood obesity., Comment: 14 pages, 9 figures, 3 tables

Since 2005 ESO has been working with its community and industry to develop an extremely large optical/infrared telescope. ESO's Extremely Large Telescope, or ELT for short, is a revolutionary ground-based telescope that will have a 39-metre main mirror and will be the largest visible and infrared light telescope in the world. To address specific topics that are needed for the science operations and calibrations of the telescope, thirteen specific working groups were created to coordinate the effort between ESO, the instrument consortia, and the wider community. We describe here the goals of these working groups as well as their achievements so far., Comment: 8 pages, 4 figures

Near infrared color centers in silicon are emerging candidates for on-chip integrated quantum emitters, optical access quantum memories and sensing. We access ensemble G color center formation dynamics and radiation-induced atomic disorder in silicon for a series of MeV proton flux conditions. Photoluminescence results reveal that the G-centers are formed more efficiently by pulsed proton irradiation than continuous wave proton irradiation. The enhanced transient excitations and dynamic annealing within nanoseconds allows optimizing the ratio of G-center formation to nonradiative defect accumulation. The G-centers preserve narrow linewidths of about 0.1 nm when they are generated by moderate pulsed proton fluences, while the linewidth broadens significantly as the pulsed proton fluence increases. This implies vacancy/interstitial clustering by overlapping collision cascades. Tracking G-center properties for a series of irradiation conditions enables sensitive probing of atomic disorder, serving as a complimentary analytical method for sensing damage accumulation. Aided by ${\it ab}$ ${\it initio}$ electronic structure calculations, we provide insight into the atomic disorder-induced inhomogeneous broadening by introducing vacancies and silicon interstitials in the vicinity of a G-center. A vacancy leads to a tensile strain and can result in either a redshift or blueshift of the G-center emission, depending on its position relative to the G-center. Meanwhile, Si interstitials lead to compressive strain, which results in a monotonic redshift. High flux and tunable ion pulses enable the exploration of fundamental dynamics of radiation-induced defects as well as methods for defect engineering and qubit synthesis for quantum information processing., Comment: 9 pages, 7 figures

In the present experiment we measure the forward-emitted angle-integrated neutron spectrum of the 7Li(p,n)7Be reaction via time-of-flight neutron spectrometry. The proton beam is shaped to a Gaussian-like distribution with energies close to the reaction threshold upstream the thick lithium target. The angle-integrated neutron spectrum resembles a Maxwellian or stellar neutron spectrum at kT=30 keV.

We present a sample of 24 local star-forming galaxies observed with broad- and narrow-band photometry from the Hubble Space Telescope, that are part of the GOALS survey of local luminous and ultra-luminous infrared galaxies. With narrow-band filters around the emission lines H$\alpha$ (and [NII]) and Pa$\beta$, we obtain robust estimates of the dust attenuation affecting the gas in each galaxy, probing higher attenuation than can be traced by the optical Balmer decrement H$\alpha$/H$\beta$ alone by a factor of $>1$ mag. We also infer the dust attenuation towards the stars via a spatially-resolved SED-fitting procedure that uses all available HST imaging filters. We use various indicators to obtain the star formation rate (SFR) per spatial bin, and find that Pa$\beta$ traces star-forming regions where the H$\alpha$ and the optical stellar continuum are heavily obscured. The dust-corrected Pa$\beta$ SFR recovers the 24$\mu$m-inferred SFR with a ratio $-0.14\pm0.32$ dex and the SFR inferred from the $8\mathrm{-}1000\,\mu\mathrm{m}$ infrared luminosity at $-0.04\pm0.23$ dex. Both in a spatially-resolved and integrated sense, rest-frame near infrared recombination lines can paint a more comprehensive picture of star formation across cosmic time, particularly with upcoming JWST observations of Paschen-series line emission in galaxies as early as the epoch of reionization., Comment: Accepted for publication in ApJS

This is the Snowmass2021 Energy Frontier (EF) Beyond the Standard Model (BSM) report. It combines the EF topical group reports of EF08 (Model-specific explorations), EF09 (More general explorations), and EF10 (Dark Matter at Colliders). The report includes a general introduction to BSM motivations and the comparative prospects for proposed future experiments for a broad range of potential BSM models and signatures, including compositeness, SUSY, leptoquarks, more general new bosons and fermions, long-lived particles, dark matter, charged-lepton flavor violation, and anomaly detection., Comment: 108 pages + 38 pages references and appendix, 37 figures, Report of the Topical Group on Beyond the Standard Model Physics at Energy Frontier for Snowmass 2021. The first nine authors are the Conveners, with Contributions from the other authors

Particle accelerators have enabled forefront research in high energy physics and other research areas for more than half a century. Accelerators have directly contributed to 26 Nobel Prizes in Physics since 1939 as well as another 20 Nobel Prizes in Chemistry, Medicine and Physics with X-rays. Although high energy physics has been the main driving force for the development of the particle accelerators, accelerator facilities have continually been expanding applications in many areas of research and technology. For instance, active areas of accelerator applications include radiotherapy to treat cancer, production of short-lived medical isotopes, synchrotron light sources, free-electron lasers, beam lithography for microcircuits, thin-film technology and radiation processing of food. Currently, the largest and most powerful accelerator is the Large Hadron Collider (LHC) at CERN, which accelerates protons to multi-TeV energies in a 27 km high-vacuum ring. To go beyond the maximum capabilities of the LHC, the next generation of circular and linear particle colliders under consideration, based on radiofrequency acceleration, will require multi-billion investment, kilometric infrastructure and a massive power consumption. These factors pose serious challenges in an increasingly resource-limited world. Therefore, it is important to look for alternative and sustainable acceleration techniques. This review article pays special attention to novel accelerator techniques to overcome present acceleration limitations towards more compact and cost-effective long term future accelerators., Comment: 11 pages, 10 figures, review report

Defect engineering is foundational to classical electronic device development and for emerging quantum devices. Here, we report on defect engineering of silicon single crystals with ion pulses from a laser accelerator with ion flux levels up to 10^22 ions/cm^2/s. Low energy ions from plasma expansion of the laser-foil target are implanted near the surface and then diffuse into silicon samples that were locally pre-heated by high energy ions. We observe low energy ion fluences of ~10^16 cm^-2, about four orders of magnitude higher than the fluence of high energy (MeV) ions. In the areas of highest energy deposition, silicon crystals exfoliate from single ion pulses. Color centers, predominantly W and G-centers, form directly in response to ion pulses without a subsequent annealing step. We find that the linewidth of G-centers increase in areas with high ion flux much more than the linewidth of W-centers, consistent with density functional theory calculations of their electronic structure. Laser ion acceleration generates aligned pulses of high and low energy ions that expand the parameter range for defect engineering and doping of semiconductors with tunable balances of ion flux, damage rates and local heating.

Plasmonic modes offer the potential to achieve PetaVolts per meter fields, that would transform the current paradigm in collider development in addition to non-collider searches in fundamental physics. PetaVolts per meter plasmonics relies on collective oscillations of the free electron Fermi gas inherent in the conduction band of materials that have a suitable combination of constituent atoms and ionic lattice structure. As the conduction band free electron density, at equilibrium, can be as high as $\rm 10^{24}cm^{-3}$, electromagnetic fields of the order of $\rm 0.1 \sqrt{\rm n_0(10^{24}cm^{-3})} ~ PVm^{-1}$ can be sustained by plasmonic modes. Engineered materials not only allow highly tunable material properties but quite critically make it possible to overcome disruptive instabilities that dominate the interactions in bulk media. Due to rapid shielding by the free electron Fermi gas, dielectric effects are strongly suppressed. Because the ionic lattice, the corresponding electronic energy bands and the free electron gas are governed by quantum mechanical effects, comparisons with plasmas are merely notional. Based on this framework, it is critical to address various challenges that underlie PetaVolts per meter plasmonics including stable excitation of plasmonic modes while accounting for their effects on the ionic lattice and the electronic energy band structure over femtosecond timescales. We summarize the ongoing theoretical and experimental efforts as well as map out strategies for the future. Extreme plasmonic fields can shape the future by not only bringing tens of TeV to multi-PeV center-of-mass-energies within reach but also by opening novel pathways in non-collider HEP. In view of this promise, we invite the scientific community to help realize the immense potential of PV/m plasmonics and call for significant expansion of the US and international R\&D program., Comment: contribution to Snowmass 2021