Archivio della ricerca - Fondazione Bruno Kessler
Not a member yet
21227 research outputs found
Sort by
DiversityOne: A Multi-Country Smartphone Sensor Dataset for Everyday Life Behavior Modeling
No full textUnderstanding everyday life behavior of young adults through personal devices, e.g., smartphones and smartwatches, is key for various applications, from enhancing the user experience in mobile apps to enabling appropriate interventions in digital health apps. Towards this goal, previous studies have relied on datasets combining passive sensor data with human-provided annotations or self-reports. However, many existing datasets are limited in scope, often focusing on specific countries primarily in the Global North, involving a small number of participants, or using a limited range of pre-processed sensors. These limitations restrict the ability to capture cross-country variations of human behavior, including the possibility of studying model generalization, and robustness. To address this gap, we introduce DiversityOne, a dataset which spans eight countries (China, Denmark, India, Italy, Mexico, Mongolia, Paraguay, and the United Kingdom) and includes data from 782 college students over four weeks. DiversityOne contains data from 26 smartphone sensor modalities and 350K+ self-reports. As of today, it is one of the largest and most diverse publicly available datasets, while featuring extensive demographic and psychosocial survey data. DiversityOne opens the possibility of studying important research problems in ubiquitous computing, particularly in domain adaptation and generalization across countries, all research areas so far largely underexplored because of the lack of adequate datasets
Spatial interactions and micro-enterprises’ uptake of COVID-19 financial aid: Evidence from a spatial hurdle probit model
No full textThis study examines the impact of spatial interactions on local government financial aid uptake among micro-enterprises (MEs) during the COVID-19 pandemic. While government support measures played a critical role in mitigating the crisis, the extent to which eligible firms applied for these benefits remains understudied. Using a representative survey of MEs in Trentino, Italy, linked to administrative data, we employ a spatial hurdle probit model to assess how information sharing among firms influences uptake decisions. Results show that MEs with similar levels of pre-pandemic economic performance in terms of added value tend to reciprocally influence each other in deciding to ask for public aid, with spatial interactions particularly significant in peripheral areas. Our findings highlight the importance of peer effects in shaping firms’ responses to public support measures, offering insights for policymakers aiming to improve the accessibility and effectiveness of financial aid programs
Towards More Effective Human-Robot Collaboration via Accurate Pose Estimation
No full textThis paper explores the pivotal role of accurate pose estimation in enhancing Human-Robot Collaboration (HRC) effectiveness. It delves into the integration of novel AI-based zero-shot estimation algorithms within collaborative robotic platforms and discusses candidate algorithms suitable for industrial scenarios
Detection of single-mode thermal microwave photons using an underdamped Josephson junction
No full textWhen measuring electromagnetic radiation of frequency f, the most sensitive detector counts single quanta of energy hf. Single photon detectors have been demonstrated from γ-rays to infrared wavelengths, with ongoing efforts to extend their range to microwaves. Here we show that an underdamped Josephson junction can detect 14 GHz thermal photons, with energy 10 yJ or 50 μeV, stochastically emitted by a microwave copper cavity at millikelvin temperatures. After characterizing the source and the detector, we vary the cavity temperature and measure the photon rate. The device achieves 45% efficiency and a dark count rate of 0.1 Hz over several GHz. Demonstrated super-Poissonian photon statistics is a signature of thermal light and a hallmark of quantum chaos. We discuss applications in dark matter axion searches and note its relevance to quantum information and fundamental physics
Promoting zirconia with carbon: Enhanced hybrid ZrO2/C catalyst for the ketonization of diluted aqueous acetic acid
No full textCO2 electroreduction (CO2ER) fuelled by renewable electricity in state-of-the-art flow-cells with solid-state electrolytes (SSE) produces aqueous solutions of C1-C2 carboxylic acids with concentrations up to 15-30 wt %, but the synthesis of longer-chain compounds remains elusive. Multicarbon e-fuels/e-chemicals could be instead obtained by upgrading acetic acid (AA) solutions produced by CO2ER with well-established processes (e.g., ketonization, consecutive aldol condensation/aromatization, hydrogenation) provided that the employed catalysts withstand water inhibition. In this work, novel hybrid catalysts based on ZrO2 deposited over a high surface area carbon support were successfully applied for the first time to the continuous-flow, gas-phase ketonization of AA to acetone (AC) with increasing amounts of steam in the feed. The catalysts were prepared by a simple incipient wetness impregnation of a commercial carbon and thoroughly characterized by N2 porosimetry, XRD, XPS, HRTEM, SEM-EDS, Raman spectroscopy, and TGA, to uncover structure-activity relationships. Steam co-feeding, even in low amounts, rendered bulk ZrO2 completely unactive due to the preferential adsorption of water, resulting in blockage of the active sites required for AA ketonization. On the other hand, the carbon support effectively hindered water adsorption on ZrO2 even with highly diluted feeds (e.g., 15 wt % of AA in water), thus playing a fundamental role in the catalytic cycle by creating a hydrophobic environment at the boundary with small (∼5 nm) and highly dispersed ZrO2 nanoparticles. The comparison with the available literature showed that the AC productivity achieved in optimized conditions with the carbon-promoted ZrO2 is the highest reported so far
Generating Counterfactual Explanations Under Temporal Constraints
No full textCounterfactual explanations are one of the prominent eXplainable Artificial Intelligence (XAI) techniques, and suggest changes to input data that could alter predictions, leading to more favourable outcomes. Existing counterfactual methods do not readily apply to temporal domains, such as that of process mining, where data take the form of traces of activities that must obey to temporal background knowledge expressing which dynamics are possible and which not. Specifically, counterfactuals generated off-the-shelf may violate the background knowledge, leading to inconsistent explanations. This work tackles this challenge by introducing a novel approach for generating temporally constrained counterfactuals, guaranteed to comply by design with background knowledge expressed in Linear Temporal Logic on process traces (LTLp). We do so by infusing automata-theoretic techniques for LTLp inside a genetic algorithm for counterfactual generation. The empirical evaluation shows that the generated counterfactuals are temporally meaningful and more interpretable for applications involving temporal dependencies
Search for the associated production of charm quarks and a Higgs boson decaying into a photon pair with the ATLAS detector
No full textA search for the production of a Higgs boson and one or more charm quarks, in which the Higgs boson decays into a photon pair, is presented. This search uses proton-proton collision data with a centre-of-mass energy of
√s = 13 TeV and an integrated luminosity of 140 fb−1 recorded by the ATLAS detector at the Large Hadron Collider. The analysis relies on the identification of charm-quark-containing jets, and adopts an approach based on Gaussian process regression to model the non-resonant di-photon background. The observed (expected, assuming the Standard Model signal) upper limit at the 95% confidence level on the cross-section for producing a Higgs boson and at least one charm-quark-containing jet that passes a fiducial selection is found to be 10.6 pb (8.8 pb). The observed (expected) measured cross-section for this process is 5.3 ± 3.2 pb (2.9 ± 3.1 pb)
On-chip purification of extracellular vesicles for microRNA biomarkers analysis
No full textExtracellular vesicles (EVs) and their cargo are increasingly suggested as innovative biomarkers correlated to the diagnosis, progression and therapy of diseases like cancer. Several techniques have been developed for the specific separation of the different classes of EVs that give solutions enriched in vesicles, but still containing other unwanted components. New methods for a more efficient, reliable and automated isolation of EVs are therefore highly desirable. Here, microparticles with surfaces endowed with positive ions were exploited to separate vesicles from complex biological matrices. First, flat silicon oxide surfaces functionalized with different divalent cations were tested for their efficiency in terms of small EV capture. Small EVs pre-purified via serial ultracentrifugations were employed for these analyses. The two better-performing cations, i.e., Cu2+ and Ni2+, were then selected to functionalize magnetic microbeads to be inserted in microfluidic chips and evaluated for their efficiency in capturing EVs and for their release of biomarkers. The best protocol setup was explored for the capture of EVs from cell culture supernatants and for the analysis of a class of biomarkers, i.e., microRNAs, via RT-PCR. The promising results obtained with this on-chip protocol evidenced the potential automation, miaturization, ease-of-use and the effective speed of the method, allowing a step forward toward its integration in simple and fast biosensors capable of analyzing the desired biomarkers present in EVs, helping the spread of biomarker analysis in both clinical settings and in research
Super Resolution of Satellite-Based Land Surface Temperature Through Airborne Thermal Imaging
No full textUrban heat island pose a significant threat to public health and urban livability. UHI maps are created using satellite thermal data, a crucial source for earth monitoring and for delivering mitigation strategies. Nowadays there is still a resolution gap between high-resolution optical data and low-resolution satellite thermal imagery. This study introduces a novel deep learning approach—named Dilated Spatio-Temporal U-Net (DST-UNet)—to bridge this gap. DST-UNET is a modified U-Net architecture which incorporates dilated convolutions to address the multiscale nature of urban thermal patterns. The model is trained to generate high-resolution, airborne-like thermal maps from available, low-resolution satellite imagery and ancillary data. Our results demonstrate that the DST-UNet can effectively generalise across different urban environments, enabling municipalities to generate detailed thermal maps with a frequency far exceeding that of traditional airborne campaigns. This framework leverages open-source data from missions like Landsat to provide a cost-effective and scalable solution for continuous, high-resolution urban thermal monitoring, empowering more effective climate resilience and public health initiatives