ICSC and Spoke 2 - 40 months later

Session

POSTER AND VIDEO UPLOAD

Not scheduled

LNF ed.36 - B. Touschek (LNF)

39. A multi-modal approach for the classification of multimessenger events involving gravitational waves and CTAO data

Physical Poster shown at the Meeting

The first multimessenger observation involving gravitational waves, GW170817, demonstrated the essential role of combining information from different messengers and from the entire electromagnetic spectrum to achieve a comprehensive understanding of astrophysical phenomena. In the coming years, the Cherenkov Telescope Array Observatory (CTAO) — the largest and most sensitive ground-based...

45. A Pipeline for Detecting Gamma-Ray Bursts in INTEGRAL SPI-ACS Data

Physical Poster shown at the Meeting

Gamma-ray bursts (GRBs), especially weak or short-duration events, are essential probes of high-energy astrophysical processes. The Anti-Coincidence Shield of the INTEGRAL/SPI instrument (SPI-ACS) provided continuous high-energy light curves well suited for detecting such transients. We developed a pipeline to search for GRBs in these data by adaptively estimating the background and...

34. A Reinforcement Learning Framework for Fraud Detection in Highly Imbalanced Financial Data

Physical Poster shown at the Meeting

Anomaly detection in financial transactions is a challenging task, primarily due to severe class imbalance and the adaptive behavior of fraudulent activities. This paper presents a reinforcement learning for fraud detection (RLFD) framework to address this problem. We train a deep Q-network (DQN) agent with a long short-term memory (LSTM) encoder to process sequences of financial events and...

37. Accelerating Geant4 Simulations through a VAE-base Super-Resolution Framework

Virtual only posters, accompanied by a 5 min video

We present the successful development and optimization of a Variational Autoencoder (VAE) framework designed to accelerate Geant4 Monte Carlo simulations in hadrontherapy applications. This work, conducted as part of ICSC Spoke 2 - Flagship 2.6.2, addresses the critical computational bottleneck in high-resolution Linear Energy Transfer (LET) calculations. Our system employs deep learning to...

29. Advanced Distributed Computing for Astrophysical Applications

Physical Poster shown at the Meeting

The rapid growth of astrophysical data from modern telescopes and numerical simulations requires scalable computational infrastructures capable of supporting high-throughput data processing and advanced modeling. Within the framework of the National Centre for HPC, Big Data, and Quantum Computing (Spoke 2), this project presents the design, deployment, and optimization of a hybrid CPU–GPU...

42. Aerial Image Analysis with YOLOv11-seg and LLaMA-4: Two Complementary Studies on a Real-World Dataset

Virtual only posters, accompanied by a 5 min video

Aerial images are difficult to analyze due to their high resolution, non-intuitive structure, and the limited availability of domain-specific datasets. We created a new real-world dataset of agricultural areas in Sicily by extracting high-resolution images from Google Maps and manually annotating eight classes with polygon masks using Roboflow. Using this dataset, we explored two complementary...

38. AI-driven analysis of multi-resolution Earth Observation data for vineyard disease pressure and crop water demand

Virtual only posters, accompanied by a 5 min video

In this abstract, we present a multi-scale framework that combines satellite and aerial imagery with machine-learning models to support advanced decision-making in Mediterranean agriculture, with a focus on vineyards. The work brings together two complementary application domains: early detection of vine diseases and irrigation management based on crop evapotranspiration (ETc).

Within the...

36. An unsupervised deep learning approach for rare signal detection in astroparticle experiments: the use-case DAIDREAM

Physical Poster shown at the Meeting

In the era of multi-messenger astronomy, where data from gamma- and cosmic-ray observatories, neutrino telescopes, gravitational interferometers, and dark matter experiments are increasingly combined, novel approaches are necessary to drive progress in astroparticle physics. The growing complexity and interconnectivity of experimental datasets demand robust solutions for data sharing,...

27. Application of Diffusion models for remote sensing super resolution

Physical Poster shown at the Meeting

This work explores the application of Denoising Diffusion Probabilistic Models (DDPMs) to super-resolution tasks in remote sensing imagery. Many research domains within Earth observation face a scarcity of high-quality data, largely due to the complexity and cost of acquiring satellite imagery. To address this challenge, we use Sentinel-2 data from the Copernicus Programme of the European...

28. Benchmarking Cloud and Bare-Metal Architectures for Interactive and Quasi‑Interactive HEP Analyses

Physical Poster shown at the Meeting

Within the framework of the ICSC project (Italian National Centre on HPC, Big Data and Quantum Computing), a flexible and experiment‑agnostic cloud infrastructure has been developed to address the increasing computational demands of the HL‑LHC and future collider experiments. The platform provides transparent access to computing resources through containerization technologies and orchestration...

21. Boosted Objects in Vector Boson Scattering: from classical to quantum ML prospects

Physical Poster shown at the Meeting

The study of Vector Boson Scattering (VBS) at LHC provides an unique window into the electroweak symmetry breaking mechanism. The polarization of the vector bosons enables precision tests of the heart Standard Model at the TeV scale, additionally sensibile
to new physics. The highest experimental sensitivity can be achieved in the boosted regime, where the bosons are produced with large...

31. Declarative paradigms for analysis description and implementation - a demonstrator for High Energy Physics

Physical Poster shown at the Meeting

The software toolbox used for big data analysis is rapidly changing in the last years. The adoption of software design approaches able to exploit the new hardware architectures and increase code expressiveness plays a pivotal role in boosting both development and performance of sustainable data analysis.

The scientific collaborations in the field of High Energy Physics (e.g. the LHC...

35. Deep learning for satellite imagery: wildfire segmentation using time-series of images from the Sentinel constellation

Physical Poster shown at the Meeting

The application of neural networks to remote sensing has already brought a transformative advancement in environmental monitoring, offering cutting-edge tools for the extraction of complex spatio-temporal patterns essential for informed environmental decision-making and rapid response to emerging crises.
This work leverages satellite observations to mainly address post-disaster assessment...

40. Deep Learning reconstruction of neutral mesons from multiple calorimetric clusters in LHCf

Virtual only posters, accompanied by a 5 min video

In this work, we introduce a deep learning–driven strategy to enhance the reconstruction of neutral meson events in the Large Hadron Collider forward (LHCf) experiment. Located in the very forward region of the LHC, LHCf measures neutral particles produced at very small angles in proton–proton and proton–ion collisions, providing crucial input for modelling hadronic interactions in ultra–high...

44. Detection and Study of Space Debris using TNG and SiFAP2 within the IDL Project

Physical Poster shown at the Meeting

In this abstract, we present the development of an integrated system with tools for Space Situational Awareness (SSA) within the Interoperable Data Lake (IDL) project. Among system's capabilities, we report a proof-of-concept study on optical debris detection using SiFAP2, an ultra-fast optical timing instrument installed at the Telescopio Nazionale Galileo (TNG).

We implemented a...

41. Development of a Machine Learning Classifier for Deuteron Identification in the GAPS Experiment

Physical Poster shown at the Meeting

The General Antiparticle Spectrometer (GAPS) is a balloon-borne experiment designed to measure low-energy cosmic-ray antinuclei below 250 MeV/n, aiming to detect indirect signatures of dark matter annihilation or decay. By covering this previously unexplored low-energy region, GAPS will achieve unprecedented sensitivity to antideuteron and antihelium fluxes. The experiment will undertake three...

26. Development of AI models for data cubes analysis

Physical Poster shown at the Meeting

The rapid growth of modern data acquisition systems has intensified the need for advanced Artificial Intelligence methods capable of processing large, heterogeneous, and multidimensional datasets. This research focuses on the development of deep learning architectures tailored for the analysis of data cubes: structures integrating spatial, spectral, and temporal information.
In this work,...

33. End-to-End Simulation Frameworks and Machine Learning Techniques for Astronomical Spectrographs and X-ray Data Exploration

Physical Poster shown at the Meeting

The growing complexity of astronomical instruments and the need for efficient, data-driven analysis require software frameworks that can fully exploit modern computing infrastructures. Within the context of the ICSC Spoke 2 - Fundamental Research & Space Economy, this work combines physics-based simulations and machine-learning techniques to support both instrument development and data-driven...

20. Exploring the portability of the ALICE ITS clustering to alpaka

Virtual only posters, accompanied by a 5 min video

This work presents the integration of the Alpaka library into the ALICE O² framework to enable portable, high-performance execution of the ITS clustering algorithm across heterogeneous hardware. By refactoring the original CPU-based code to a Struct of Arrays (SoA) layout and implementing an Alpaka kernel for masked-pixel filtering, we achieved consistent results and up to **20%...

24. GWTBoost: Exploring Frequency-Dependent Anisotropies in the Gravitational-Wave Detector Network Sensitivity

Virtual only posters, accompanied by a 5 min video

The sensitivity of the global gravitational-wave detector network is known to vary across the sky. In this study, part of the GWTboost non-flagship use case and utilising the LEONARDO high-performance computing resources, we investigate the global directional sensitivity of the current LIGO–Virgo–KAGRA network during the O4a and O4b observing runs. To this end, we study the network’s...

25. GWTBoost: Modernizing Unmodeled Gravitational-Wave Analysis from cWB to PycWB in the High-Performance Computing Era

Virtual only posters, accompanied by a 5 min video

Unmodeled search methods play a crucial role in detecting generic gravitational-wave transients (GWTs), especially in the case of signals without precise theoretical predictions. Coherent WaveBurst (cWB) is one of the primary data-analysis pipelines used for the detection and coherent reconstruction of such signals. Within the GWTBoost non-flagship use case, we focus on developing and...

22. Integration of ICSC provisioned resources hosted at INFN Naples into the high rate platform via InterLink

Virtual only posters, accompanied by a 5 min video

The High Rate Analysis platform that has been implemented in the context of the Working Group 5 of ICSC Spoke2 offers a general purpose environment where analyzers can scale up their computations via the python Dask library. This can be done either distributing the workload within the actual Kubernetes cluster resources, or offloading to remote resources. The latter option is enabled by the...

30. Low latency, fast inference neural networks on FPGA: the ATLAS L0 Muon Barrel Trigger use case

Physical Poster shown at the Meeting

In many research and industrial settings, achieving fast, low-latency algorithmic responses is essential. To meet the demands of the upgraded LHC and future High Energy Physics (HEP) detectors, quick and powerful triggers are necessary. In recent years, Machine Learning (ML) algorithms have been widely applied to such tasks, and more recently, hardware solutions for ML applications based on...

23. ML-Based Cluster Counting for Particle Identification

Physical Poster shown at the Meeting

Cluster counting is a highly promising particle identification technique for drift chambers in particle physics experiments.. In this paper, we trained neural network models, including a Long Short-Term Memory (LSTM) model for the peak-finding algorithm and a Convolutional Neural Network (CNN) model for the clusterization algorithm, using various hyperparameters such as loss functions,...

43. POD pipeline optimization and containerization

Physical Poster shown at the Meeting

It is here described a development activity in the context of ICSC Spoke 2, WP3, use case UC 2.3.6, "Data processing pipeline optimization for space and ground based experiments", and named "Pipeline for space gravity missions". The main objective is the development of methodologies and expertise for the optimization and management of pipelines for precise orbit determination (POD) of...

48. POSTER AND VIDEO UPLOAD

46. The Frequency-Hough project for continuous gravitational waves: algorithm optimization and scalability to HPC systems.

Physical Poster shown at the Meeting

The search for continuous and persistent gravitational waves emitted by isolated and rotating neutron stars is a top priority for current and future ground-based detectors. However, those searches are typically bounded in sensitivity by their high computational costs. In this work, we will show the optimization work done on the Frequency-Hough algorithm, which performs a blind search for...

32. Updates on WP1 projects from the theory group at Universita' della Calabria

Virtual only posters, accompanied by a 5 min video

In this talk I will present the status and advances of different projects related to WP1, which are ongoing within the theory group at Universita' della Calabria.