Archivio della ricerca - Fondazione Bruno Kessler
Not a member yet
21227 research outputs found
Sort by
Abstraction Modulo Stability
No full textThe analysis of legacy systems requires the automated extraction of high-level specifications. We propose a framework, called Abstraction Modulo Stability, for the analysis of transition systems operating in stable states, and responding with run-to-completion transactions to external stimuli. The abstraction captures, in the form of a finite state machine, the effects of external stimuli on the system state. This approach is parametric on a set of predicates of interest and on the definition of stability. We consider some possible stability definitions, which yield different practically relevant abstractions, and propose parametric algorithms for abstraction computation. The framework is evaluated in terms of expressivity and adequacy within an industrial project with the Italian Railway Network, on reverse engineering of relay-based interlocking circuits to extract specifications for a computer-based reimplementation
Wild Berry image dataset collected in Finnish forests and peatlands using drones
No full textBerry picking has long-standing traditions in Finland, yet it is challenging and can potentially be dangerous. The integration of drones equipped with advanced imaging techniques represents a transformative leap forward, optimising harvests and promising sustainable practices. We propose WildBe, the first image dataset of wild berries captured in peatlands and under the canopy of Finnish forests using drones. Unlike previous and related datasets, WildBe includes new varieties of berries, such as bilberries, cloudberries, lingonberries, and crowberries, captured under severe light variations and in cluttered environments. WildBe features 3,516 images, including a total of 18,468 annotated bounding boxes. We carry out a comprehensive analysis of WildBe using six popular object detectors, assessing their effectiveness in berry detection across different forest regions and camera types. We will release WildBe publicly
CRISPR/Cas9 screens identify LIG1 as a sensitizer of PARP inhibitors in castration-resistant prostate cancer
No full textPARP inhibitors (PARPi) have received regulatory approval for the treatment of several tumors, including prostate cancer (PCa), and demonstrate remarkable results in the treatment of castration-resistant prostate cancer (CRPC) patients characterized by defects in homologous recombination repair (HRR) genes. Preclinical studies showed that DNA repair genes (DRG) other than HRR genes may have therapeutic value in the context of PARPi. To this end, we performed multiple CRISPR/Cas9 screens in PCa cell lines using a custom sgRNA library targeting DRG combined with PARPi treatment. We identified LIG1, EME1, and FAAP24 losses as PARPi sensitizers and assessed their frequencies from 3 to 6% among CRPC patients. We showed that concomitant inactivation of LIG1 and PARP induced replication stress and DNA double-strand breaks, ultimately leading to apoptosis. This synthetic lethality (SL) is conserved across multiple tumor types (e.g., lung, breast, and colorectal), and its applicability might be extended to LIG1-functional tumors through a pharmacological combinatorial approach. Importantly, the sensitivity of LIG1-deficient cells to PARPi was confirmed in vivo. Altogether, our results argue for the relevance of determining the status of LIG1, and potentially other non-HRR DRG for CRPC patient stratification and provide evidence to expand their therapeutic options
Search for same-charge top-quark pair production in pp collisions at √s = 13 TeV with the ATLAS detector
No full textGANzzle++: Generative approaches for jigsaw puzzle solving as local to global assignment in latent spatial representations
No full textJigsaw puzzles are a popular and enjoyable pastime that humans can easily solve, even with many pieces. However, solving a jigsaw is a combinatorial problem, and the space of possible solutions is exponential in the number of pieces, intractable for pairwise solutions. In contrast to the classical pairwise local matching of pieces based on edge heuristics, we estimate an approximate solution image, i.e., a mental image, of the puzzle and exploit it to guide the placement of pieces as a piece-to-global assignment problem. Therefore, from unordered pieces, we consider conditioned generation approaches, including Generative Adversarial Networks (GAN) models, Slot Attention (SA) and Vision Transformers (ViT), to recover the solution image. Given the generated solution representation, we cast the jigsaw solving as a 1-to-1 assignment matching problem using Hungarian attention, which places pieces in corresponding positions in the global solution estimate. Results show that the newly proposed GANzzle-SA and GANzzle-VIT benefit from the early fusion strategy where pieces are jointly compressed and gathered for global structure recovery. A single deep learning model generalizes to puzzles of different sizes and improves the performances by a large margin. Evaluated on PuzzleCelebA and PuzzleWikiArts, our approaches bridge the gap of deep learning strategies with respect to optimization-based classic puzzle solvers
Search for diphoton resonances in the 66 to 110 GeV mass range using pp collisions at √s = 13 TeV with the ATLAS detector
No full textsearch is performed for light, spin-0 bosons decaying into two photons in the 66 to 110 GeV mass range, using 140 fb−1 of proton-proton collisions at √s = 13 TeV produced by the Large Hadron Collider and collected by the ATLAS detector. Multivariate analysis techniques are used to define event categories that improve the sensitivity to new resonances beyond the Standard Model. A model-independent search for a generic spin-0 particle and a model-dependent search for an additional low-mass Higgs boson are performed in the diphoton invariant mass spectrum. No significant excess is observed in either search. Mass-dependent upper limits at the 95% confidence level are set in the model-independent scenario on the fiducial cross-section times branching ratio into two photons in the range of 8 fb to 53 fb. Similarly, in the model-dependent scenario upper limits are set on the total cross-section times branching ratio into two photons as a function of the Higgs boson mass in the range of 19 fb to 102 fb
Soil moisture forecasting from sensors-based soil moisture, weather and irrigation observations: A systematic review
No full textAgriculture is one of the most essential industries since it provides food for the entire population worldwide. Maintaining limited water resources is a challenging problem in this field, as growing healthy vegetables and fruits require consistent plants watering. To automatize this maintenance, software companies started developing solutions utilizing artificial intelligence tools to forecast soil moisture levels from past observations of soil humidity, weather and irrigation, measured by different sensors. This forecast is useful for irrigation decisions support and crop growth monitoring. Even though such solutions are widely developed, still, a transparent, unified methodology how forecasting models for irrigation management from sensors should be designed and evaluated is still missing. In this paper, we provide such methodology from analysis of state-of-the-art scientific articles presenting forecasting methods for soil moisture from sensor data. This review tackles several research question of how to forecast future soil moisture level from sensor-based past observations of soil moisture, weather and irrigation information. Furthermore, we follow the standard Preferred Reporting Items for Systematic Reviews and Meta-Analyses procedures for literature search analysis in computer science. As a result of literature search, we summarized 60 scientific articles presenting soil moisture forecast published from 2014 to 2024. In conclusion, we present the main challenges in forecasting soil moisture and suggest how they can be addressed
The Venus score for the assessment of the quality and trustworthiness of biomedical datasets
No full textBiomedical datasets are the mainstays of computational biology and health informatics projects, and can be found on multiple data platforms online or obtained from wet-lab biologists and physicians. The quality and the trustworthiness of these datasets, however, can sometimes be poor, producing bad results in turn, which can harm patients and data subjects. To address this problem, policy-makers, researchers, and consortia have proposed diverse regulations, guidelines, and scores to assess the quality and increase the reliability of datasets. Although generally useful, however, they are often incomplete and impractical. The guidelines of Datasheets for Datasets, in particular, are too numerous; the requirements of the Kaggle Dataset Usability Score focus on non-scientific requisites (for example, including a cover image); and the European Union Artificial Intelligence Act (EU AI Act) sets forth sparse and general data governance requirements, which we tailored to datasets for biomedical AI. Against this backdrop, we introduce our new Venus score to assess the data quality and trustworthiness of biomedical datasets. Our score ranges from 0 to 10 and consists of ten questions that anyone developing a bioinformatics, medical informatics, or cheminformatics dataset should answer before the release. In this study, we first describe the EU AI Act, Datasheets for Datasets, and the Kaggle Dataset Usability Score, presenting their requirements and their drawbacks. To do so, we reverse-engineer the weights of the influential Kaggle Score for the first time and report them in this study. We distill the most important data governance requirements into ten questions tailored to the biomedical domain, comprising the Venus score. We apply the Venus score to twelve datasets from multiple subdomains, including electronic health records, medical imaging, microarray and bulk RNA-seq gene expression, cheminformatics, physiologic electrogram signals, and medical text. Analyzing the results, we surface fine-grained strengths and weaknesses of popular datasets, as well as aggregate trends. Most notably, we find a widespread tendency to gloss over sources of data inaccuracy and noise, which may hinder the reliable exploitation of data and, consequently, research results. Overall, our results confirm the applicability and utility of the Venus score to assess the trustworthiness of biomedical data
A decade of gender bias in machine translation
No full textGender bias in machine translation (MT) has been studied for over a decade, a time marked by societal, linguistic, and technological shifts. With the early optimism for a quick solution in mind, we review over 100 studies on the topic and uncover a more complex reality—one that resists a simple technical fix. While we identify key trends and advancements, persistent gaps remain. We argue that there is no simple technical solution to bias. Building on insights from our review, we examine the growing prominence of large language models and discuss the challenges and opportunities they present in the context of gender bias and translation. By doing so, we hope to inspire future work in the field to break with past limitations and to be less focused on a technical fix; more user-centric, multilingual, and multiculturally diverse; more personalized; and better grounded in real-world needs
RGCVAE: relational graph conditioned variational autoencoder for molecule design
No full textIdentifying molecules that exhibit some pre-specified properties is a difficult problem to solve. In the last few years, deep generative models have been used for molecule genera- tion. Deep Graph Variational Autoencoders are among the most powerful machine learning tools with which it is possible to address this problem. However, existing methods strug- gle to capture the true data distribution and tend to be computationally expensive. In this work, we propose RGCVAE, an efficient and effective Graph Variational Autoencoder based on: (i) an encoding network exploiting a new powerful Relational Graph Isomor- phism Network; (ii) a novel probabilistic decoding component. Compared to several State- of-the-Art VAE methods on two widely adopted datasets, RGCVAE shows State-of-the-Art molecule generation performance while being significantly faster to train. The Python code implementing RGCVAE is openly accessible for download at: https://github.com/drigoni/ RGCVAE