Research
SDEGnO – SDE on GPU for Optimization
Ca’ Foscari University of Venice and INFN – National Institute for Nuclear Physics, Milan-Bicocca Division, Milan and Venice, Italy
Actively involved in a research collaboration focused on the development of high-performance, general-purpose simulators for the simulation and automatic calibration of physical models based on stochastic differential equations (SDEs). The project employs Monte Carlo tech- niques and global optimization algorithms—including evolutionary strategies, parameter cali- bration methods, and swarm intelligence approaches—to significantly reduce simulation time and energy consumption, while maintaining a high level of scientific accuracy. The work takes full advantage of NVIDIA multi-GPU architectures and High Performance Computing (HPC) technologies, aiming to generate impactful contributions to the scientific community through future publications.
Swarm Intelligence for Optimization and Machine Learning
NOVA IMS, Universidade Nova de Lisboa, Lisbon, Portugal
Engaged in an international research collaboration focused on the study and refinement of Swarm Intelligence algorithms for optimization and machine learning tasks.
ALLIANCE Project – Precision Delivery of Nucleic Acid-Based Therapeutics
Ca’ Foscari University of Venice, Italy
Active involvement in the research project “ALLIANCE – A Novel Integrated Cyclic Peptide- Based Platform for Precision Delivery of Nucleic Acid-Based Therapeutics,” funded under Italy’s PNRR (Mission 4, Component 2 – Investment 1.4), within the National Center for Gene Therapy and Drugs based on RNA Technology, supported by NextGenerationEU. The project aims to overcome the limitations of nucleic acid-based therapeutics (DNA/RNA) by designing innovative nanocarriers—functionally similar to miniaturized antibodies—that can selectively bind to specific cell surface receptors. These systems promise enhanced tissue penetration, reduced toxicity, and improved therapeutic efficacy. Responsibilities include the development of evolutionary algorithms and computational (in sil- ico) strategies for the design and optimization of cyclic peptides with high affinity and specificity for targeted protein binding. Research activities rely on High Performance Computing (HPC) resources with the goal of identifying peptide candidates for advanced biomedical applications.
Cardio-Imaging and Clinical Decision Support
University Hospital – University of Padua, Department of Radiology, Padua, Italy
Collaboration on a research project aimed at integrating quantitative and qualitative data anal- ysis techniques in the field of cardio-imaging, with the goal of developing innovative tools to support clinical decision-making. The research combines advanced methodologies including Machine Learning, Artificial Intelligence (AI), Explainable AI (XAI), and Fuzzy Inference Sys- tems to build interpretable and reliable models that assist healthcare professionals in cardiac image analysis and patient assessment. The main objective is to enhance the understanding of decision-making processes in clinical settings by identifying meaningful data patterns, reducing diagnostic uncertainty, and propos- ing predictive, explainable tools. This approach supports the effectiveness of medical deci- sions while maintaining the central role of the clinician and fostering a constructive interaction between human expertise and computational support.
Optimization of Fuzzy Self-Tuning PSO (FST-PSO) for Swarm Intelligence Applications
Univesity of Trieste and Ca’ Foscari University of Venice, Italy
The project aims to optimize and enhance the FST-PSO algorithm, a swarm intelligence-based global optimization technique to solve real- and discrete-valued multi-dimensional minimization problems. The project refines the adaptive mechanism of the FST-PSO, utilizing fuzzy logic for dynamic tuning of key parameters, such as cognitive, social and inertia factors, toward an improved particle behavior and convergence.
MASTER Project: Mobility Data Analysis in Public Transportation
Ca’ Foscari University of Venice, in collaboration with ACTV S.p.A., Venice, Italy
Collaboration within the European project “MASTER – Multiple ASpects TrajEctoRy manage- ment and analysis,” as part of a Data Science research initiative focused on mobility data anal- ysis in the context of public transportation in Venice. The project, developed in partnership with ACTV S.p.A., the public transport operator of Venice, aimed to enhance the efficiency and sustainability of the city’s transport system, addressing its unique urban and logistical challenges. Research activities focused on the application of Mobility Data Mining and clustering techniques for the analysis of large-scale ticket validation datasets.