Calls 2024

The field of machine learning and computational intelligence has seen significant advancements in the development of models for regression problems. However, there is a growing demand for models that not only provide accurate predictions but also offer interpretability and trustworthiness, especially in domains where decision-makers require a clear understanding of the underlying reasoning process. This research aims to explore and develop evolving fuzzy systems that address these challenges, focusing on regression problems.

We are looking for PhD students which are interested in exploring causality aspects of Deep learning and Artificial Intelligence.

This is a very trending topic in the Computer Science and Deep Learning community and impacts several research areas:
explainable AI, recommender systems, formal logic, generative AI, automated learning, etc. Examples of aspects to be investigated are:

- Deep Learning and predictive methods applied to temporal data
- Causality aspects of learning policies and their evaluation
- Knowledge Representation and Linear Time Temporal Logic

This topic is particularly interesting because it can benefit from expertise coming from different disciplines: computer science, physics, mathematics, neurobiology. Therefore, students with a strong scientific background in one (or more) of these subjects are encouraged to apply.

Moreover, the research performed through this scholarship can range from abstract and foundational research to very applied one and therefore the topic can adapt to the candidate's natural inclination and preferences.

The field of machine learning and computational intelligence has seen significant advancements in the development of models for regression problems. However, there is a growing demand for models that not only provide accurate predictions but also offer interpretability and trustworthiness, especially in domains where decision-makers require a clear understanding of the underlying reasoning process. This research aims to explore and develop evolving fuzzy systems that address these challenges, focusing on regression problems.

The proposed topic of the thesis is related to ultra-thin PCBs, tailored for applications where intricate designs demand cutting-edge space optimization, such as in satellite payloads or large-scale scientific experiments. In detector systems, minimizing PCB thickness is often necessary to reduce dead material in the active region, where the sensor is highly sensitive to any perturbations. This is crucial for both space-based and ground-based scientific experiments.

The PhD candidate will undertake a comprehensive study encompassing (i) design and simulation, (ii) manufacturing and (iii) experimental campaigns for the high-frequency characterization (up to 30 GHz) of custom Printed Circuit Boards (PCBs) and various bonding schemes to chip-to-flex interconnections.

Ultra-thin PCBs will either be manufactured in FBK via custom patent-pending techniques or by commercial standards to be used as a benchmark. The candidate will design and simulate the PCB stack, including differential pairs and controlled impedance routing.

Full process control during manufacturing will enhance the model development, allowing for the identification of specific contributions from the macroscopic geometric features (such as the shape of the metal leads) to microscopic elements like lead roughness, grain size (see Mayadas-Shatzkes model) and bonding types.

The study will explore various bonding techniques, including wire-bonding, TAB bonding and bump bonding for 3D integration. These techniques differ in materials and bonding geometries, affecting impedance and signal insertion loss. Thus, developing a computational model (e.g.using Comsol) and validating it with experimental measurements is critical for selecting the appropriate electronics design.

By validating the simulated data with VNA measurements, the investigation aims to deepen the understanding on how each factor included in the model influences the signal integrity of PCBs in high-frequency applications. Those insights will inform the design of advanced assemblies for scientific detectors in future experiments at CERN or in space missions conducted by ASI, ESA and NASA.

The proposed topic of the thesis is related to ultra-thin PCBs, tailored for applications where intricate designs demand cutting-edge space optimization, such as in satellite payloads or large-scale scientific experiments. In detector systems, minimizing PCB thickness is often necessary to reduce dead material in the active region, where the sensor is highly sensitive to any perturbations. This is crucial for both space-based and ground-based scientific experiments.

The PhD candidate will undertake a comprehensive study encompassing (i) design and simulation, (ii) manufacturing and (iii) experimental campaigns for the high-frequency characterization (up to 30 GHz) of custom Printed Circuit Boards (PCBs) and various bonding schemes to chip-to-flex interconnections.

Ultra-thin PCBs will either be manufactured in FBK via custom patent-pending techniques or by commercial standards to be used as a benchmark. The candidate will design and simulate the PCB stack, including differential pairs and controlled impedance routing.

Full process control during manufacturing will enhance the model development, allowing for the identification of specific contributions from the macroscopic geometric features (such as the shape of the metal leads) to microscopic elements like lead roughness, grain size (see Mayadas-Shatzkes model) and bonding types.

The study will explore various bonding techniques, including wire-bonding, TAB bonding and bump bonding for 3D integration. These techniques differ in materials and bonding geometries, affecting impedance and signal insertion loss. Thus, developing a computational model (e.g.using Comsol) and validating it with experimental measurements is critical for selecting the appropriate electronics design.

By validating the simulated data with VNA measurements, the investigation aims to deepen the understanding on how each factor included in the model influences the signal integrity of PCBs in high-frequency applications. Those insights will inform the design of advanced assemblies for scientific detectors in future experiments at CERN or in space missions conducted by ASI, ESA and NASA.

Microfluidics is a branch of microsystem technologies aiming at integrating complex analytical procedures into miniaturized, automated devices. In this kind of systems, precise handling of tiny liquid samples is of paramount importance to enable the detection of specific biological and chemical species.

The overall goal of this research project is to develop novel microfluidic devices based on the technology of microelectromechanical systems (MEMS) to achieve miniaturization and integration of different functionalities, including sample management and analysis. To this aim, Electrowetting On Dielectrics (EWOD) techniques will be explored. In EWOD devices the sample and reagents are managed in the form of drops, which are moved, split and mixed by modulating the wetting properties of surfaces by electric fields.

Research project and expected outcomes The candidate will participate in a MEMS research team developing microfluidic devices, with access to full MEMS design and microfabrication facilities. The research project will focus on the design and modeling of novel microfluidic devices, the microfabrication of the MEMS devices in a fully equipped, class 100 cleanroom, and the experimental validation of the obtained prototypes. The results of the research activities will be collected in contributions submitted to international scientific journals, and presented in high profile conferences.

Microfluidics is a branch of microsystem with great potential to integrate complex analytical procedures into a small device, which allows the detection of biological and chemical analytes in portable and automated systems. Among different techniques, the recent trend is to move from classical but miniaturized pressure-driven devices to droplet techniques and digital microfluidics, which allow a better miniaturization and integration of different functionalities without an increase of the technological complexity. In particular, the Electrowetting On Dielectrics (EWOD) is a droplet-based technique, where the sample and reagents are managed by modulating the wetting properties of a surface by electric fields.

The candidate will participate in a research team developing EWOD devices for portable analytical systems for the analysis of water quality and other applications. The activities will include the development of MEMS technologies, the design and the experimental validation of the devices.

This scholarship aims to pioneer the development of open-source robotics to create cost-effective agricultural digital twins, bridging the gap between digital and physical farming systems. By leveraging the versatility and accessibility of open-source platforms, it will focus on designing, developing, and implementing robotic systems capable of accurately simulating agricultural environments. These digital twins will enable farmers to predict crop outcomes, optimize resource allocation, and mitigate risks by providing a virtual representation of their fields, thus facilitating informed decision-making processes. Key objectives include the development of scalable and modular robotic platforms that can be customized to suit diverse agricultural needs, the integration of advanced sensors and AI algorithms for real-time data processing and simulation, and the establishment of a framework for the seamless transition between digital twins and their physical counterparts. The project will also explore innovative approaches to reduce costs and improve the accessibility of robotics in agriculture, making cutting-edge technology available to a wider range of users.

The development of low-cost, efficient, printable RFID sensors is a fundamental research domain for the Internet of Things (IoT). Chipless RFIDs are a new and emerging technology that removes the silicon chip from the sensor tag, including both the identification and the sensing function in the tag design. A chipless sensor is made by one or more RF resonant structures, whose frequency is dependent on the dielectric material covering the metallic resonator. This PhD position is about the study of chipless RFID technology for the sensing of environmental parameters. The study is interdisciplinary and will have different combined approaches: -At material level: study and identify the best polymer formulation for maximising the sensitivity to of parameters. At device level: build a sensor prototype with the most promising materials, for multi-parameter detection. Evaluate its performances with RF measurements, establishing also the suitability for wireless detection. The work will be primarily in FBK in collaboration with DII-UniTN.

Silicon Carbide offers unique physical properties which can be exploited for high-performance new radiation detectors. This activity aims at developing radiation sensors with three-dimensional electrodes (either columnaror trench-shaped) in Silicon Carbide for applications in harsh environments. The PhD research program will be focused on the design, TCAD simulation, material and fabrication aspects, as well as experimental characterization of prototypes.

Research focus will be on the development and experimental validation of digital twins for hybrid power system (HPS) and hybrid energy storage systems (HESS). Deployment of HPS has been growing lately, such systems require advanced control and forecasting for optimal management. Digital twins based on hybrid models have the potential to improve system operation increasing reliability and resilience but further advancements are needed.

This grant is funded by project IPCEI Batterie 2 — CUP B62C22000010001.

This scholarship aims to pioneer the development of open-source robotics to create cost-effective agricultural digital twins, bridging the gap between digital and physical farming systems. By leveraging the versatility and accessibility of open-source platforms, it will focus on designing, developing, and implementing robotic systems capable of accurately simulating agricultural environments. These digital twins will enable farmers to predict crop outcomes, optimize resource allocation, and mitigate risks by providing a virtual representation of their fields, thus facilitating informed decision-making processes. Key objectives include the development of scalable and modular robotic platforms that can be customized to suit diverse agricultural needs, the integration of advanced sensors and AI algorithms for real-time data processing and simulation, and the establishment of a framework for the seamless transition between digital twins and their physical counterparts. The project will also explore innovative approaches to reduce costs and improve the accessibility of robotics in agriculture, making cutting-edge technology available to a wider range of users.

The development of low-cost, efficient, printable RFID sensors is a fundamental research domain for Internet of Things (IoT). Chipless RFIDs are a new and emerging technology that removes the silicon chip from the sensor tag, including both the identification and the sensing function in the tag design.
A chipless sensor is basically made by one or more RF resonant structures, whose frequency is dependent on the dielectric material covering the metallic resonator.
This PhD position is about the study of chipless RFID technology for the sensing of ionizing radiation. The study will have different combined approaches:
-At material level: study and identify the best polymer formulation for maximising the sensitivity to different types of radiation (x-rays, ?-rays)
-At device level: build a sensor prototype with the most promising covering material, for both high-dose and low-dose irradiation.
Evaluate its performances, with radiation test and RF measurements, establishing also the suitability for wireless detection.
The work will be in collaboration between DII-Unitn and FBK, with irradiation tests performed at TIFPA or in other Italian INFN labs

Digital twins are dynamic and self-evolving models that simulate a physical asset and represent its exact state through bi-directional data assimilation. They employ Artificial Intelligence (AI) data-driven and symbolic techniques to provide state synchronization, monitoring, control and decision support.

This project will build on the results of the ongoing ESA-funded project ExploDTwin (""Digital Twin for Space Exploration Assets""), which aims at integrating the DT into the ESA infrastructure to support online space assets operations with functionalities such as planning, what-if-analysis, fault detection, diagnosis and prognosis. To this end, ExploDT introduces a model-based design methodology that allows to seamlessly integrate engineering methods and AI techniques into a cohesive DT framework.

The PhD student will investigate new formal methods to analyze the DTs by integrating model checking, automated theorem proving, simulation, and machine learning. Different aspects of the DTs will be considered including temporal properties for validation, monitorability and diagnosability. The new methods will be implemented and evaluated on space related benchmarks derived from ExploDTwin results.

This initiative contributes to global food security and environmental sustainability, fostering a collaborative and innovative research atmosphere focused on integrating data science and artificial intelligence (AI) with agriculture. It addresses urgent challenges such as resource management, crop yield optimization, and the impacts of climate change. As the global demand for food increases, this scholariship promotes the adoption and creation of technologies and methodologies to boost agricultural efficiency and sustainability. The successful candidate will engage in developing innovative, data-driven methods for sustainable agricultural practices. Additionally, he/she will explore interdisciplinary research, utilizing remote sensing, machine learning, and predictive analytics to produce actionable insights aimed at enhancing agricultural productivity and resilience. This scholarship aims to cultivate a new breed of researcher, adept at leveraging data intelligence for agricultural progress, with a focus on optimizing water usage, enhancing crop resilience, and implementing sustainable farming practices.

In the context of the European Space Agency (ESA) mission JUpiter ICy moons Explorer (JUICE) to the Jovian system, and the development of the ESA EnVision mission to Venus, we seek for candidates willing do develop methodologies for radar sounder data processing (data enhancement, semantic segmentation, denoising, content based retrieval, target detection, multitemporal analysis, etc.). The outcome of this activity will contribute in improving the understanding of the subsurface of planetary bodies, the correlation to history and climate as well as to a better understanding of the Earth.

The candidate will be requested to design and develop novel methodologies within artificial intelligence framework (machine learning, deep learning, pattern recognition, etc.) for effective information extraction from radar sounder data.

Besides the requirements established by the rules of the ICT school, preferential characteristics for candidates for this scholarship are:

•    master degree in Electrical Engineering, Communication Engineering, Computer/Data Science, Mathematics or equivalents;

•    knowledge in artificial intelligence, image/signal processing, remote sensing, radar remote sensing.

This scholarship is funded by project ASI-INAF n. 2023-6-HH.0 JUICE-RIME-E - “Missione JUICE - Attività dei team scientifici dei Payload per Lancio, commissioning, operazioni e analisi dati” — CUP no. F83C23000070005

Nowadays, foundation models have shown remarkable capabilities in generating text, images and videos, rich world knowledge, and some complex “reasoning” skills. However, these models are still passive models (e.g., action-oriented aspects of intelligence are not leveraged), static, data- and computation-expensive, they often confabulate, and are difficult to align with human values. This PhD project aims at making multimodal foundation models capable of interactions with other agents (e.g., enabling cooperative capabilities) and of interactions grounded in the world, enabling counterfactual reasoning and causality abilities in foundation models as well as enforcing their alignment with human intentions and values. The PhD candidate is required to have previous experience in working with deep learning algorithms, a strong interest on transformers’ architecture, graph neural networks, multimodality, and cooperative and embodied AI. The selected student will be able to collaborate with the ELLIS network and being part of the ELLIS PhD program (if selected) as well as with top universities and research centers such as MIT, Max Planck, etc.

Techniques based on formal methods for the verification and validation of embedded and safety-critical software systems are becoming increasingly important, due to the growing complexity and importance of such systems in every aspect of modern society. Despite the major progress seen in the last twenty years, however, the application of formal methods in embedded software remains a challenge in practice, due to factors such as the interplay between computation and physical aspects and the increasing complexity of the software and its configurations. This project will investigate novel techniques for the application of formal methods to the design, verification, and validation of embedded software, with particular emphasis on safety-critical application domains such as railways, automotive, avionics, and aerospace. The techniques considered will include a combination of automated and interactive theorem proving, satisfiability modulo theories, model checking, abstract interpretation, and deductive verification. Examples of the problems tackled during the project include the formal verification of functional requirements expressed in temporal logics, automated test-case generation, efficient handling of parametric/multi-configuration software systems and product lines, and the verification of software operating in a physical environment, subject to real-time constraints. Importantly, in addition to researching novel theoretical results, a significant part of the project activities will be devoted to the implementation of the techniques in state-of-the-art verification tools developed at FBK and their application to real-world problems in collaboration with our industrial partners.

Industrial systems are reaching an unprecedented degree of complexity. The process of designing a complex system is expensive, time consuming and error-prone. Moreover, the design process has to guarantee not only the functional correctness of the implemented system, but also its dependability and resilience with respect to run-time faults. Hence, the design process must characterize the likelihood of faults, mitigate possible failures, and assess the effectiveness of the adopted mitigation measures. Formal methods have been increasingly used over the last decades to deal with the shortcomings of designing a complex system. Formal methods are based on the adoption of a formal, mathematical model of the system, shared between all actors involved in the system design, and on a tool-supported methodology to aid all the steps of the design, from the definition of the architecture down to the final implementation in HW and SW. Formal methods include technologies such as model checking, an automatic technique to symbolically and exhaustively analyze all possible executions of the system in the formal model, in order to detect design flaws as early as possible. Model checking techniques have been recently extended to assess the safety and dependability characteristics of the design, and for system certification. The objective of this study is to advance the state-of-the-art in system design using formal methods. This includes adapting and extending the system design methodology, investigating improved versions of state-of-the-art routines for verification and safety assessment of complex systems, and developing novel extensions to address open problems. Examples of such extensions include novel techniques for contract-based design and contract-based safety assessment, advanced techniques for formal verification based on compositional reasoning, the analysis of the timing aspects of fault propagation, the characterization of transient and sporadic faults, the analysis of the effectiveness of fault mitigation measures in presence of complex fault patterns, and the modeling of analysis of systems with continuous and hybrid dynamics. This study will exploit the challenges and benchmarks defined in various industrial projects carried out at FBK.

The advent of foundation models such as large language models (LLMs) has led to impressive advancements in all areas of natural language processing. Exploiting their capabilities for many downstream applications has therefore emerged as an interesting research direction, as well as extending them to other modalities. Toward this aim, researchers have started investigating the combination of speech encoders and LLMs. Despite the potential of this emerging approach, systematic studies are still required to isolate its strengths and weaknesses compared to traditional systems, as well as to identify the most effective strategies for specific applications, spanning from architectural choices to selecting the optimal data and tasks for training the adapter between the speech encoder and the LLM. In this scenario, the PhD candidate will study existing solutions, identify their weaknesses, and develop innovative approaches to overcome them, contributing to the advancement of the field, either on a theoretical basis, with an application-oriented focus, or both.

The Internet of Things (IoT) paradigm proposes massive increases in multimodal data creation on tiny devices located at the far edges of the computing infrastructure. Despite the resource constraints of these devices, communication constraints require the data they produce to be locally processed before transmission. Further, due to the increasing complexity of the data generated, machine learning techniques are being successfully applied at the far edge through so-called TinyML. While most TinyML techniques focus on the individual device, increasingly these devices are both networked together and/or connected to larger cloud or swarm based infrastructures, introducing challenges for managing communication and coordination, both in data collection and usage as well as regarding elements of distributed training and continuous learning. Research into these challenges requires innovation in i) applying novel techniques such as distillation, hardware aware scaling and neural architecture search to implement orchestratable TinyML algorithms, ii) innovative communication, including optimizing the low power communication network connecting edge devices to the infrastructure, and iii) orchestration techniques required for ML-enhanced edge devices to participate in complex distributed applications formed of heterogeneous devices. This interdisciplinary research at the intersection of Artificial Intelligence, Embedded Systems, Distributed Computing, and Low-power Hardware will take into account the candidate's profile and interests, contributing to the development of innovative solutions for real-world challenges. The candidate will have the opportunity to work with cutting-edge technology, gain valuable experience in interdisciplinary collaboration, and make significant contributions to the field of machine learning at the very edge.

An Urban Digital Twin is a dynamic digital model of a city, fed by data collected from the city itself, and capable of faithfully reproducing the city behaviour through the use of advanced modelling, data science, and AI techniques. One of the key foreseen applications of Urban Digital Twins is to predict the evolution of the city, including the effects and impacts of external changes (eg, climate change) and internal processes (eg, urban transition policities and incentives). A challenge for the development of Urban Digital Twins is that cities are systems-of-systems, with complex interactions between physical, organisational, and social dimensions, as well as between the physical world and the digital world. Novel modeling and simulation techniques are necessary to develop Urban Digital Twins able to manage this complexity and produce reliable predictions.

The candidate will be requested to work in this research area, and in particular to work on advanced agent-based modeling and simulation frameworks for Urban Digital Twins and to their validation on real scenarios concerning the adoption of Digital Twin by Italian cities. Besides the requirements established by the rules of the ICT school, preferential characteristics for candidates for this scholarship are:

• master degree in Computer/Data Science, Mathematics, Phisics, Electrical Engineering, Communication Engineering, or equivalents;

• knowledge in artificial intelligence, complex systems, agent-based modeling and simulation.

The continuous release of new and increasingly powerful language models is opening possibilities to address applicative scenarios that were not even imaginable a few years ago. In particular, the goal of this PhD is to improve strategic and tactical communication activities during crisis events, through the use of advanced natural language generation and persuasive communication techniques. More specifically, the idea is to directly intervene with textual responses and narratives that are meant to address the public during and after crisis events and natural disasters. To this end the candidate should focus on all those aspects of the LLM, such as decoding strategies, knowledge guided generation, data quality, knowledge distillation, reinforcement learning from human feedback -just to mention a few- that can help in improving the models, especially for better factuality, reducing hallucination and increasing coherence among messages while assisting professionals in crisis management.

Publicly available multimodal datasets often consist merely of text captions paired with images, text with audio, or audio with images, while also merely providing descriptions of the images without establishing any deep-level connection between the text content and the visual or auditory content. Such formats fail to reflect the complexity of human perception, such as how we process images, listen to audio, or comprehend text. The PhD candidate will be tasked with exploring new modalities that more accurately reflect human vision and attention as they align with verbal descriptions. This initiative aims to foster a profound understanding of how visually perceived content (including images, videos, and text) is processed and understood by humans. Achieving this understanding will allow us to advance in the comprehension capabilities of Large Multimodal Models, while enhancing their ability to interpret and interact with a wide range of data.

3D scene understanding is an area of vision research with applications ranging from augmented reality to autonomous navigation. This PhD position is focused on research into foundational models for 3D scene understanding. 3D scenes can be created from image collections (Structure from Motion, Simultaneous Localisation and Mapping), thus allowing for the extraction of foundational representations from images and their transfer to the 3D domain using pixel-to-point correspondences. These representations can then be interacted with via language model prompting. However, these representations have been optimised for 2D reasoning. The PhD candidate will be tasked with exploring novel approaches to disentangle object-level information in the 2D domain and to fuse it in 3D, thereby enabling 3D reasoning capabilities.

The Internet of Things (IoT) paradigm proposes massive increases in multimodal data creation on tiny devices located at the far edges of the computing infrastructure. Despite the resource constraints of these devices, communication constraints require the data they produce to be locally processed before transmission. Further, due to the increasing complexity of the data generated, machine learning techniques are being successfully applied at the far edge through so-called TinyML. While most TinyML techniques focus on the individual device, increasingly these devices are both networked together and/or connected to larger cloud or swarm based infrastructures, introducing challenges for managing communication and coordination, both in data collection and usage as well as regarding elements of distributed training and continuous learning. Research into these challenges requires innovation in i) applying novel techniques such as distillation, hardware aware scaling and neural architecture search to implement orchestratable TinyML algorithms, ii) innovative communication, including optimizing the low power communication network connecting edge devices to the infrastructure, and iii) orchestration techniques required for ML-enhanced edge devices to participate in complex distributed applications formed of heterogeneous devices.

This interdisciplinary research at the intersection of Artificial Intelligence, Embedded Systems, Distributed Computing, and Low-power Hardware will take into account the candidate's profile and interests, contributing to the development of innovative solutions for real-world challenges. The candidate will have the opportunity to work with cutting-edge technology, gain valuable experience in interdisciplinary collaboration, and make significant contributions to the field of machine learning at the very edge.

In the context of the European Space Agency (ESA) mission JUpiter ICy moons Explorer (JUICE) to the Jovian system, and the development of the ESA EnVision mission to Venus, we seek for candidates willing do develop methodologies for radar sounder data processing (data enhancement, semantic segmentation, denoising, content based retrieval, target detection, multitemporal analysis, etc.). The outcome of this activity will contribute in improving the understanding of the subsurface of planetary bodies, the correlation to history and climate as well as to a better understanding of the Earth.

The candidate will be requested to design and develop novel methodologies within artificial intelligence framework (machine learning, deep learning, pattern recognition, etc.) for effective information extraction from radar sounder data.

Besides the requirements established by the rules of the ICT school, preferential characteristics for candidates for this scholarship are:

• master degree in Electrical Engineering, Communication Engineering, Computer/Data Science, Mathematics or equivalents;

• knowledge in artificial intelligence, image/signal processing, remote sensing, radar remote sensing.

This scholarship is funded by project ASI-INAF n. 2023-6-HH.0 JUICE-RIME-E - “Missione JUICE - Attività dei team scientifici dei Payload per Lancio, commissioning, operazioni e analisi dati” — CUP no. F83C23000070005

The recent Earth Observation missions like (ESA Copernicus - Sentinels, ASI PRISMA and COSMO-SkyMed, and future IRIDE constellation) make available databases of long, dense and worldwide image time series. The data have complex spatio-spectro-temporal behaviors and variability, and they show irregularities and misalignments.

Candidates will be requested to develop novel methodologies within artificial intelligence framework (machine learning, deep learning, pattern recognition, etc.) for effectively and efficiently process image time series for semantic segmentation, target detection and change detection across multiannual series of data.

Besides the requirements established by the rules of the ICT school, preferential characteristics for candidates for this scholarship are:

• master degree in Electrical Engineering, Communication Engineering, Computer/Data Science, Mathematics or equivalents;

• background in artificial intelligence, image/signal processing, remote sensing, passive/active sensors.

A Network Intrusion Detection System (NIDS) serves as the initial line of defence against network attacks that threaten the integrity of data, systems, and networks. Over recent years, Machine Learning (ML) algorithms have been increasingly used in NIDSs to detect malicious traffic due to their remarkable accuracy in identifying malicious network activity.

Nevertheless, ML algorithms are susceptible to Adversarial Machine Learning (AML) attacks, which aim to evade the NIDS with small perturbations of the attack network traffic. This vulnerability has particularly severe consequences, as adversarial attacks pose a substantial threat to overall network security.

While the majority of current research in the field of AML has been directed towards computer vision tasks like image classification and object recognition, there has been a notable increase in interest and activity within the cybersecurity domain. Nevertheless, several challenges persist in this domain, encompassing both performance-related issues and the practicality of applying these methods to real-world scenarios.

The primary objective of this PhD scholarship is to conduct cutting-edge research in the field of AML with a focus on enhancing cybersecurity defences. The selected candidate will explore innovative techniques and methodologies to detect, prevent, and mitigate AML attacks, thereby improving the robustness and resilience of ML-based cybersecurity systems.

The drive to establish a cloud-to-edge continuum, spanning multiple heterogeneous compute regions, is progressively eroding the boundaries of security perimeters, calling for a zero-trust approach to security, where nothing, not even inside an organisation’s network, can be implicitly trusted. In such a scenario, a resource- and energy-efficient, opportunistic monitoring of users, services, platform, and infrastructure becomes paramount to enhance management and security in cloud-to-edge environments.

Monitoring and auditing have received increasing attention from research and industry in the recent years. Crucially, novel technologies have emerged to programmatically gather information from network flows, system calls, and other sources. These technologies, including eBPF (extended Berkeley Packet Filter) and P4, fall under the umbrella of the so-called programmable data planes.

The objective of this PhD endeavour is to design, implement and evaluate novel monitoring solutions capable of opportunistically diving into the appropriate depth of data collection, defining the right mix of data and user/control plane functions and their location. Additionally, they should be programmatically tailored to serve security and data analysis applications, while being suitable for dynamic scaling and orchestration. The overarching goal is to contain their footprint while delivering the required information effectively.

Programmable Data Planes (PDPs) offer the ability to customise and control the processing of network packets, either within network devices such as routers, switches, and SmartNICs, or within end-host machines, through technologies like eBPF (extended Berkeley Packet Filter), or other programmable frameworks. PDPs empower developers to define and implement customised packet processing logic, spanning from fundamental packet filtering and forwarding to more sophisticated tasks like load balancing, network virtualisation, and security enforcement.

While they promise to deliver enhanced flexibility and performance to processes and services, the capabilities of PDPs are still largely under exploration, proof, and assessment. As a matter of facts, due to design and implementation decisions, they are often restricted in the range and types of operations they can perform on packets. This thesis aims to delve into novel and advanced methodologies for offloading complex tasks, particularly those focused on security such as cryptography, traffic analysis, and filtering, onto PDPs, with the objective of striking the right balance between performance and complexity. The candidate is expected to analyse alternative solutions where tasks can be either entirely or partially offloaded and conduct experimental assessments, comparing the outcomes against legacy approaches.

This initiative contributes to global food security and environmental sustainability, fostering a collaborative and innovative research atmosphere focused on integrating data science and artificial intelligence (AI) with agriculture. It addresses urgent challenges such as resource management, crop yield optimization, and the impacts of climate change. As the global demand for food increases, this scholariship promotes the adoption and creation of technologies and methodologies to boost agricultural efficiency and sustainability. The successful candidate will engage in developing innovative, data-driven methods for sustainable agricultural practices. Additionally, he/she will explore interdisciplinary research, utilizing remote sensing, machine learning, and predictive analytics to produce actionable insights aimed at enhancing agricultural productivity and resilience. This scholarship aims to cultivate a new breed of researcher, adept at leveraging data intelligence for agricultural progress, with a focus on optimizing water usage, enhancing crop resilience, and implementing sustainable farming practices.

The proliferation of Internet of Things (IoT) devices and the increasing reliance on edge computing paradigms have ushered in a new era of distributed computing, where processing occurs closer to the data source rather than in centralized data centers. This shift promises to reduce latency, minimize bandwidth usage, and ensure data privacy and sovereignty. However, it also introduces significant challenges in managing, deploying, and maintaining machine learning (ML) models across a vast, heterogeneous, and geographically dispersed infrastructure. This doctoral research aims to address these challenges by advancing the integration of Machine Learning Operations (MLOps) practices within edge computing and IoT environments, facilitating the seamless deployment, monitoring, and management of ML models at the edge.

Techniques based on formal methods for the verification and validation of embedded and safety-critical software systems are becoming increasingly important, due to the growing complexity and importance of such systems in every aspect of modern society. Despite the major progress seen in the last twenty years, however, the application of formal methods in embedded software remains a challenge in practice, due to factors such as the interplay between computation and physical aspects and the increasing complexity of the software and its configurations.

This project will investigate novel techniques for the application of formal methods to the design, verification, and validation of embedded software, with particular emphasis on safety-critical application domains such as railways, automotive, avionics, and aerospace. The techniques considered will include a combination of automated and interactive theorem proving, satisfiability modulo theories, model checking, abstract interpretation, and deductive verification. Examples of the problems tackled during the project include the formal verification of functional requirements expressed in temporal logics, automated test-case generation, efficient handling of parametric/multi-configuration software systems and product lines, and the verification of software operating in a physical environment, subject to real-time constraints. Importantly, in addition to researching novel theoretical results, a significant part of the project activities will be devoted to the implementation of the techniques in state-of-the-art verification tools developed at FBK and their application to real-world problems in collaboration with our industrial partners.

Industrial systems are reaching an unprecedented degree of complexity. The process of designing a complex system is expensive, time consuming and error-prone. Moreover, the design process has to guarantee not only the functional correctness of the implemented system, but also its dependability and resilience with respect to run-time faults. Hence, the design process must characterize the likelihood of faults, mitigate possible failures, and assess the effectiveness of the adopted mitigation measures.

Formal methods have been increasingly used over the last decades to deal with the shortcomings of designing a complex system. Formal methods are based on the adoption of a formal, mathematical model of the system, shared between all actors involved in the system design, and on a tool-supported methodology to aid all the steps of the design, from the definition of the architecture down to the final implementation in HW and SW. Formal methods include technologies such as model checking, an automatic technique to symbolically and exhaustively analyze all possible executions of the system in the formal model, in order to detect design flaws as early as possible. Model checking techniques have been recently extended to assess the safety and dependability characteristics of the design, and for system certification.

The objective of this study is to advance the state-of-the-art in system design using formal methods. This includes adapting and extending the system design methodology, investigating improved versions of state-of-the-art routines for verification and safety assessment of complex systems, and developing novel extensions to address open problems. Examples of such extensions include novel techniques for contract-based design and contract-based safety assessment, advanced techniques for formal verification based on compositional reasoning, the analysis of the timing aspects of fault propagation, the characterization of transient and sporadic faults, the analysis of the effectiveness of fault mitigation measures in presence of complex fault patterns, and the modeling of analysis of systems with continuous and hybrid dynamics.

This study will exploit the challenges and benchmarks defined in various industrial projects carried out at FBK.

Nowadays, foundation models have shown remarkable capabilities in generating text, images and videos, rich world knowledge, and some complex “reasoning” skills. However, these models are still passive models (e.g., action-oriented aspects of intelligence are not leveraged), static, data- and computation-expensive, they often confabulate, and are difficult to align with human values.

This PhD project aims at making multimodal foundation models capable of interactions with other agents (e.g., enabling cooperative capabilities) and of interactions grounded in the world, enabling counterfactual reasoning and causality abilities in foundation models as well as enforcing their alignment with human intentions and values. The PhD candidate is required to have previous experience in working with deep learning algorithms, a strong interest on transformers’ architecture, graph neural networks, multimodality, and cooperative and embodied AI. The selected student will be able to collaborate with the ELLIS network and being part of the ELLIS PhD program (if selected) as well as with top universities and research centers such as MIT, Max Planck, etc.

Predicting individual and collective human behavior is crucial to address complex societal challenges. Recent research has focused on deep learning models for forecasting future behavior. While these models achieve impressive results, they face limitations:

limited generalizability, low interpretability, and difficulties in geographic transfer. This PhD project aims to design the next generation of computational models for understanding individual and collective human behavior. Social science research on social learning,

collective intelligence, and crowd wisdom identifies potentially generalizable behavioral patterns. Additionally, recent advancements in AI offer foundation models capable of reasoning and generalization.

The ideal candidate possesses a strong interest in a multidisciplinary approach encompassing machine learning (deep learning and foundation models), social sciences, urban mobility, and related fields. The ultimate goal is to contribute to the development of the first foundation model for human behavior.

The collaborative nature of the project fosters engagement with leading national and international universities, creating a dynamic research environment.

Digital twins are dynamic and self-evolving models that simulate a physical asset and represent its exact state through bi-directional data assimilation. They employ Artificial Intelligence (AI) data-driven and symbolic techniques to provide state synchronization, monitoring, control and decision support.

This project will build on the results of the ongoing ESA-funded project ExploDTwin ("Digital Twin for Space Exploration Assets"), which aims at integrating the DT into the ESA infrastructure to support online space assets operations with functionalities such as planning, what-if-analysis, fault detection, diagnosis and prognosis. To this end, ExploDT introduces a model-based design methodology that allows to seamlessly integrate engineering methods and AI techniques into a cohesive DT framework.

The PhD student will investigate new formal methods to analyze the DTs by integrating model checking, automated theorem proving, simulation, and machine learning. Different aspects of the DTs will be considered including temporal properties for validation, monitorability and diagnosability. The new methods will be implemented and evaluated on space related benchmarks derived from ExploDTwin results.

Point clouds are nowadays an indispensable tool in the heritage field, but their actual usage by non-experts engaged with preservation and restoration challenges is often very limited, due to low explainability, human-readability, accessibility and data integrability issues. All of these issues fall under the conceptual umbrella of “understandability”.

AI-based approaches particularly suffer these problems, but, as point clouds, they are an indispensable tool for digital heritage. Other approaches, e.g. based on 3D ontologies, could support understanding aspects but they have been limited addressed.

Therefore the goals of the proposed PhD are:

(i) To study, develop and validate generalisable ontology-based approaches to facilitate the query and use of large and complex 3D heritage point clouds by means of rules able to infer properties and characteristics of a surveyed scene

(ii) To conduct research on novel ways to integrate formal ontologies and AI-based methods to support explainability

(iii) To integrate LLM and NLP models to support 3D heritage understanding

The successful candidate is supposed to have a good ability to connect ICT/AI solutions with heritage needs, along with the agility to successfully prototype innovative, reliable and replicable software solutions.

Biodiversity is under threat from habitat loss and fragmentation, climate change, extreme land use and invasive alien species. Biodiversity monitoring for conservation purposes using geospatial data has seen some progress in recent years. Traditional survey methods are time-consuming, labour-intensive and require skilled staff. Surveying can be challenging, especially to keep track of rare or elusive species living in inaccessible or dangerous areas. Photogrammetric and LiDAR data has led to important advancements and impacts on ecosystem understanding as they enable precise assessments of habitat structures, mapping of species distributions or ecosystem dynamics, all essential information for conservation efforts. Nonetheless, better objective processes and monitoring solutions, with new tools to enhance and enrich current practices and assist ecologists, are needed.

Therefore, the goal of the interdisciplinary PhD is to:

(i) create and validate processes, based on 3D remote sensing data (airborne/drone monochromatic/multispectral LiDAR, photogrammetric point clouds, aerial/drone hyperspectral images, etc.) and AI methods, to locate and study various species, either animal or vegetal

(ii) use ground robotics platforms and sensors for terrestrial biodiversity exploration and monitoring in challenging environments

(iii) find biodiversity patterns, through multimodal data analytics, detecting hotspots of current issues, their trends and emerging threats

(v) combine/fuse 3D data to cross-validate the methods

(vi) find new links within geographical data between animals and vegetation.

In recent years, Multi-Modal Learning (MML) has garnered significant attention, driven by the increasing availability of vast multimodal datasets and the development of robust internet services accessible across various devices. Current research predominantly emphasizes the imperative to harness deep learning techniques, capitalizing on existing foundational models applicable across diverse domains, and customizing them for specific tasks within the MML framework.

The primary objective of this thesis is to advance the state of the art in MML by delving into cutting-edge neural architectures and learning approaches. This involves the exploration of innovative methods, potentially combining different modalities such as voice and gestures. The aim is to enhance the performance of existing single-modal systems, particularly in Automatic Speech Recognition (ASR) and Natural Language Processing (NLP).

A pivotal focus will be on investigating recent audio foundation models, specifically those designed for multi-lingual speech recognition, voice conversion, and speech generation to facilitate data augmentation. The overarching goal is to develop models that exhibit effectiveness across various speech tasks, even when confronted with limited data or constrained computation resources. Furthermore, through the integration of advanced audio generation techniques, the study seeks to bolster the multimodal capabilities of the overall system. This enhancement enables more effective creation of synthetic data for model training and development.

The outcomes of this research are anticipated to contribute significantly to the development of highly competitive services tailored to the demands of evolving multimodal application scenarios.

The advent of foundation models has led to impressive advancements in all areas of natural language processing. However, their huge size poses limitations due to the significant computational costs associated with their use or adaptation. When applying them to specific tasks, fundamental questions arise: do we actually need all the architectural complexity of large and - by design - general-purpose foundation models? Can we optimize them to achieve higher efficiency? These questions spark interest in research aimed at reducing models’ size, or deploying efficient decoding strategies, so as to accomplish the same tasks while maintaining or even improving performance. Success in this direction would lead to significant practical and economic benefits (e.g., lower adaptation costs, the possibility of local deployment on small-sized hardware devices), as well as advantages from an environmental impact perspective towards sustainable AI. Focusing on automatic translation, this PhD aims to understand the functioning dynamics of general-purpose massive foundation models and explore possibilities to streamline them for specific tasks. Possible areas of interest range from textual and speech translation (e.g., how to streamline a massively multilingual model to best handle a subset of languages?) to scenarios where the latency is a critical factor, such as in simultaneous/streaming translation (e.g., how to streamline the model to reduce latency?), to automatic subtitling of audiovisual content (e.g., how to streamline the model without losing its ability to generate compact outputs suitable for subtitling?).

The growing complexity of software systems requires the development of new methods and tools to design and test software systems characterized by high variability in the space of possible functional configurations and possible release architectures. The objective of this doctoral thesis is to explore new approaches to testing, verification and validation of this type of complex systems involving the joint use of model-based and artificial intelligence techniques such as optimization, planning and machine learning.

Created as an extension of the DevOps methodology, DevSecOps adds security constraints to cloud-native application deployment and lifecycle maintenance. By fostering teamwork and automating checks, it speeds up delivery while keeping software safer, ensuring the use of security best practices in each phase of the software development.

Cloud-native applications are fundamentally a new approach to designing and building scalable software based on microservices that run on dynamic environments such as public, private, and hybrid clouds. This new approach raises a completely new set of security challenges, not only concerning the software itself but also its deployment and maintenance. Issues such as insecure cloud configuration, container orchestration mishaps, or insecure secrets storage are few examples of the main security threats mentioned in the OWASP Cloud-Native Application Security Top 10 [1].

In the context of cloud-native applications development, the candidate will be asked to explore the validation and verification of a set of security constraints in the container lifecycle, starting from container image creation and going through the entire execution phase with the goal of identifying  vulnerabilities that could make the microservice or its dependable infrastructure insecure by deploying a validator to scan the microservice code and configuration, periodically searching for vulnerabilities and other potential security threats. For this, the candidate will explore several techniques, including Natural Language Processing (NLP) and other machine learning algorithms. The candidate will also consider the application's overall security by implementing a component, integrated into CI/CD pipeline, able to fix the code, recreate the container image, or reconfigure it whenever a threat is identified.

References

[1] https://owasp.org/www-project-cloud-native-application-security-top-10

We are seeking highly motivated and ambitious candidates to join our research team and pursue an industrial PhD in collaboration with the Motivational Digital Systems research unit of Fondazione Bruno Kessler and AWorld.

Aworld is the official platform of ACTNOW, the United Nations Campaign aiming for individual action on climate change and sustainability. Therefore, the PhD candidate will have the unique opportunity to make an impact with their research to awareness and behavioral change related to environmental sustainability.

The PhD Grant endeavors to explore the transformative potential of Artificial Intelligence (AI) applied to motivational gamified systems, with a particular focus on raising individuals' awareness of environmental sustainability and promoting positive long-term behavioural change.

In recent years, the rapid evolution of AI models and techniques has presented unprecedented opportunities for enhancing personalised and immersive gaming experiences. Motivational gamified systems, designed with educational, awareness-raising, or behavioural change objectives in mind, stand to benefit significantly from the integration of AI-driven capabilities.

This PhD project aims at harnessing AI to revolutionise player profiling, content generation, adaptation, and personalization within gamified systems and experience. By leveraging AI technologies, motivational gamified systems can dynamically adapt to individual player profiles (e.g., preferences, performances, skills, motivational factors),  and fosteri deeper engagement and immersion.

The PhD Grant aims to address the following key objectives:

• Methodologies for pro-environmental gamified systems design: define methodologies and design principles for pro-environmental gamified systems that enable sustained user engagement over time (retention), studying methods to prevent fatigue and maintain high interest.

• AI-driven Game Analytics and Impact Assessment: study and develop gamification analytics to analyse user system usage, monitor progress, measure impact achieved, and evaluate motivational factors influencing user behavior.

• AI-Driven User Profiling and Content Generation: leverage AI-based techniques for user profiling and the creation of highly personalised experiences. This includes personalization of motivational elements such as storytelling, challenges, rewards, and feedback, based on individual user profiles and historical data for enhanced engagement and motivation.

• Experimental studies and validation: implement the above-mentioned AI-driven algorithms and tools aimed at promoting eco-friendly behaviors, raising awareness about environmental issues (such as climate change, pollution, conservation of resources), and encouraging action, and validate them within the AWorld platform.

Through interdisciplinary research, the PhD Grant aims to push the boundaries of AI-driven Game User Research, leading to the creation of a new generation of motivational gamified systems that are not only entertaining but also educational, impactful, and deeply immersive.

Making data-driven decisions is widely accepted among private companies. Identifying relevant data for making appropriate business decisions is crucial. High-quality data is essential for effectively using advanced techniques in decision-making processes, such as data analytics and machine learning. Despite the consensus on the data-driven approach in the private sector, the broader context of society (including organizations, groups, informal communities, and citizens) lacks an adequate technological infrastructure to share and use data securely and reliably in a privacy-preserving manner while ensuring that appropriate data quality requirements are satisfied. To address these and related issues, the European Commission initiated the creation of Common European Data Spaces (CEDS), a type of legally regulated complex ecosystems, in various strategic areas (such as healthcare and agriculture) to achieve the key objectives identified in Europe’s political, economic, and social strategy.

CEDSs aim to enable reliable and secure data that can be exchanged across the EU, allowing public and private sector operators to control and share the data they generate while integrating processes for data quality management to allow for the creation of innovative services based on advanced data processing techniques and, most prominently, AI algorithms.  This vision requires the development of a technology infrastructure that supports CEDSs and related data processes by integrating cloud and edge computing with networking, protected by security, privacy, trust, and data quality management services to enable the development and maintenance of “fit for purpose" data processes. The cornerstone of all such services is the adoption of digital identity management solutions based on the European Digital Identity (EUDI) Wallet to permit identification, authentication, and authorization of data access, auditing, and data lineage/provenance activities within and among CEDS.

By considering the complex security concerns in this domain, the candidate will be asked to develop security testing methodologies tailored to the specific challenges posed by EUDI wallets when used in different use case scenarios including, in particular, finance and banking. More concretely, the candidate will focus  on researching and implementing robust testing frameworks and techniques tailored to payment service provider apps and integrated banking app software development kits. This research contributes to enhancing the resilience of digital wallet ecosystems, securing identities, and improving trust in digital identity management practices.

Electrodes, including a Catalyst Layer (CL) and a Porous Transport Layer (PTL), have an important role in determining the performance and durability of a low-temperature electrolysis cell, making their optimization of primary importance. The composition and microstructure of the anodic CL promoting the OER, that is the rate determining step in PEM electrolysis, is particularly critical since it defines the overall cell efficiency and durability. At SoA, the anodic catalyst is mainly IrOx, which is deposited on the membrane, with the Catalyst Coated Membrane (CCM) method.

This work aims at the development of innovative electrodes with the CCS method, depositing the catalyst on the PTL. The first part of the work will focus on literature review, to better understand the state of the art of OER catalysts for PEM electrolysis. The work will then focus on the optimization of catalysts, inks and processes for CCS preparation. Small electrodes (25 cm2) will be developed in FBK and UFI laboratory facilities. The catalyst will be synthesized with innovative techniques, such as Physical Vapor Deposition (PVD) technique. Particular attention will be paid to the reduction of the loading of PGM (Platinum Group Material).

The electrodes will be first characterized by a morphological and electrochemical point of view, to assess its activity and stability. Finally, selected electrodes will be tested in real operating conditions, to assess their performance in an electrolysis cell.

PTLs have an important role in low-temperature electrolysis cells, being responsible for water/electrolyte distribution, gas (oxygen and hydrogen) removal, charge transfer, and mechanical support for the membrane. Therefore, their structural properties play an important role in defining the performance of the electrolysis cells, leading to the need of optimizing their porosity, pore diameters, pore structure, morphology, thickness, and permeability.

Thus, this work focuses on the optimization of the PTL properties and adopting both innovative methods for PTL productions, with focus on production methods such us tape casting and more established methods like tape casting or innovative ones.

The first part of the works will focus on literature review, to better understand the state of the art of PTL production and characterizations. Then, the work will focus on PTL structural optimization (for instance including flow field). The activities will include development and characterization of small size PTLs (25 cm2) in FBK and UFI laboratory facilities, as well as modelling and simulation of PTL to optimize their mass transfer, charge transfer and structural properties. Additionally, particular attention will be dedicated to the interactions at the interface between the PTL and the CL.

The second part of the work will focus on the optimization of the production methods, optimizing the production process for scale- production. This phase will consider the optimization of the process on both a technical and economic point of view.

The aim of this research project is to develop a system which, inserted into a television automation platform, is capable to automatically identify the highlights of an input video.

The goal of this project is to be able to generate and then automatically publish the result on VOD/OTT platforms (I.E.: Website, APP, HBBTV, Social Platforms).

Highlights generally mean those passages of a video which, once identified, are able to describe the most salient moments of the original video.

The system must be able to recognize the highlights based on video topology. In case of "sport-football", it will therefore have to recognize, for example, yellow and red cards as well as goals, while in “sport-tennis” it will have to be able to recognize the end of a set and the salient events based on the sentiment of the video.

For contents for which it is not possible to give a precise subcategory, it is therefore foreseen to recognize the salient passages based only on the sentiment of the video. Example, during a television talk show, recognize when the discussion becomes more heated.

The analysis carried out by the AI system will have to assign scores to the extrapolated sequences. The score indicates how important that particular segment is, in relation with the complete video, so as to allow the system that analyzes the result to be able to choose the best highlights.

Space systems have reached an unprecedented degree of complexity. The design process has to guarantee not only the functional correctness of the implemented system, but also its dependability and resilience with respect to run-time faults. Hence, the design process must characterize the likelihood of faults, mitigate possible failures, and assess the effectiveness of the adopted mitigation measures.

Formal methods have been increasingly used over the last decades to deal with the shortcomings of designing complex systems, in different domains. Formal methods are based on the adoption of a formal, mathematical model of the system, shared between all actors involved in the system design, and on a tool-supported methodology to aid all the steps of the design, from the definition of the architecture down to the final implementation in HW and SW.

The objective of this study is to advance the state-of-the-art in space system design using formal methods. In particular, it will investigate new techniques for model-based system and software engineering, to support the design, mission preparation and operations of space systems. The potential research directions include fault detection, isolation, and recovery (FDIR) for satellites; system level diagnosis and diagnosability based on telemetry; digital twins for satellites. Topics to be investigated include techniques for contract-based design and contract-based safety assessment, advanced verification techniques based on compositional reasoning, the analysis of the timing aspects of fault propagation, the characterization of transient and sporadic faults, the analysis of the effectiveness of fault mitigation measures in presence of complex fault patterns, and the modeling and analysis of systems with continuous and hybrid dynamics.

The developed techniques will be implemented and evaluated using tools for system-software engineering such as the COMPASS tool and the COMPASTA tool, based on the TASTE tool chain. The topics of the PhD are aligned with the AIFDIR (Design, Verification and Validation of AI-based FDIR) and ExploDTwin (Digital Twin for Space Exploration Assets) projects, funded by the Italian and European Space Agencies.

Single-photon solid-state sensors, such as Single Photon Avalanche Diodes (SPADs) and Silicon Photomultipliers (SiPMs), are gradually replacing the traditional Photomultiplier in several applications, ranging from big scientific experiments at CERN, to space missions, to Positron Emission Tomography to automotive LiDAR. On the other hand, an important factor limiting the use of SIPMs and SPADs in space, as well as in HEP, is the qualification and improvement of their radiation tolerance. To this end, Fondazione Bruno Kessler (FBK) is dedicating significant research effort to the study of the radiation damage in SiPMs, through experimental characterization of their characteristics after irradiation, and to the development of innovative structures featuring enhanced radiation tolerance, which will enable a whole set of new applications in space. In this context, the PhD candidate will carry out her/his research in FBK, which is one of the worldwide recognized leaders in the development of SiPMs and SPADs. The activity will focus on the experimental characterization of the electro optical properties of the devices after irradiation with both ionizing and non-ionizing particles, employing and possibly improving the advanced setups and analysis software already available at FBK and exploiting different irradiation facilities that FBK has access to, providing sources of protons, neutrons or x-rays. The PhD candidate will also study innovative detector structures with enhanced radiation tolerance, contributing to their development and becoming proficient in the understanding, characterization and use of silicon detectors for science and industrial applications. Furthermore, she/he will investigate solutions to increase the lifetime and preserve the performances of detectors in space through optimized packaging. The activity will benefit from the strong scientific network built by FBK in the space sector, including GSSI, several partners from INFN, and ESA. The research will be carried out in the context of different, space-focused projects, in which FBK is partner or coordinator, with the role of developing single-photon sensors optimized for space applications: ASTRA, funded by Italian PNRR and coordinated by GSSI, SpaceItUp, funded by ASI / PNRR, and ORBITS, funded by ESA and coordinated by FBK.

In the information age, the policy-making process struggles to effectively integrate findings from data science. Despite the abundance of data, the complexity of concepts and the difficulty in translating complex analyses into concrete policies hinder the full utilization of data. This issue is compounded by the misalignment between short-term political priorities and evidence supported by the data. Furthermore, additional data is often needed to comprehensively frame the problems at hand. Given these obstacles, a key research question is how to effectively communicate findings from data science in policy-making and ensure decisions are relevant and based on solid evidence.

The goal of this Ph.D project is to explore innovative, technology-based approaches to facilitate education for collaborative and student-centered learning environments. The project aims to develop semi-automatic and adaptive educational paths, as well as personalized support systems that benefit both educators and learners. This will involve utilizing various technologies and methods, such as end-user programming or multimodal, competence-based learning, to co-create learning activities that consider specific needs and preferences; of particular interests (but not limited to) are multilingual and speech-based tools.

The ideal candidate have a background in Computer Science, Psychology or Cognitive Science, with knowledge of data science, artificial intelligence. Experience in designing interactive digital technologies, good skills in Python programming, and quantitative analysis are strongly required. Knowledge of educational theories is a plus, and in case it has to be acquired during the initial phase of the project.

The PhD position is offered in co-tutoring between the i3 research unit of the Augmented Intelligence center of FBK and the Department of Psychology and Cognitive Science.

Accessing advanced acceleration structures with microfabrication techniques: Particle accelerators are becoming huge - the proposed Future Circular Collider with 100km circumference. Here, acceleration is addressed from the opposite end, using microfabrication to obtain extremely compact structures, to produce particle beams for a variety of possible applications. In other words: particle acceleration on a chip.

The topic will study and design possible devices. It will then utilise the micro- and nano- fabrication facilities at the Sensors&Devices to realise them and to test these prototype structures. Consideration of potential applications of such devices will be considered from the beginning of the design.

An interest and aptitude for design and microfabrication of silicon devices is expected. A hands on and practical mentality is desired. Experience of data handling and analysis is similarly helpful. An interest in the particle acceleration and applications of particle beams is useful.

Development of single or few-qubit gates for integrated quantum photonic circuits. The PhD will address the fundamnetal theoretical aspects, design, fabrication and characterisation of integrated chip-scale devices.

The LHC experiments produce about 90 petabytes of data per year. Inferring the nature of particles produced in high-energy collisions is crucial for probing the Standard Model with greater precision and searching for phenomena beyond the Standard Model. In this context, event selection is becoming more difficult than ever before and requires expertise at the border between physics and computer science. During the PhD the student will be guided in exploring and designing algorithms that mix background knowledge with Deep Learning to tackle this problem, learning to apply rigorous Data Science methodologies. The activity will be carried out in collaboration with INFN-TIFPA, Fondazione Bruno Kessler, and within the ATLAS experiment at the LHC. Candidates familiar with High Energy Physics are welcome, and basic knowledge of Machine Learning/Deep Learning is recommended.

Silicon Carbide (SiC) has emerged as a promising semiconductor material for electronic devices due to its exceptional material properties, including high thermal conductivity, wide bandgap, and excellent chemical and mechanical stability. As SiC-based devices become increasingly integral to various applications, understanding and optimizing dopant characteristics is crucial for advancing their performance and reliability. This thesis focuses on the comprehensive characterization of dopants in SiC semiconductors, aiming to bridge the gap between material science and device engineering. The research involves the exploration of various dopant types, concentrations, and their impact on the electrical, structural, and thermal properties of SiC. Both traditional and novel doping techniques are investigated, with an emphasis on their effects on carrier mobility, doping efficiency, and device functionality. The experimental methodology encompasses advanced analytical techniques such as Secondary Ion Mass Spectrometry (SIMS), X-ray Photoelectron Spectroscopy (XPS), and Hall effect measurements to precisely quantify and analyze dopant profiles, dopant activation levels, and their distribution within the SiC lattice. Additionally, the impact of dopants on defect formation and migration in SiC will be investigated to gain insights into the material's long-term stability and reliability. The research also explores the practical implications of dopant characterization on SiC device performance. Through the integration of the acquired knowledge, this study aims to propose optimized doping strategies for enhancing the efficiency and reliability of SiC-based electronic devices. The findings of this research hold significant potential for advancing the field of SiC semiconductors, enabling the development of next-generation electronic components with improved performance, efficiency, and longevity.

Tip-enhanced Raman Spectroscopy (TERS) combines the chemical analysis of the Raman technique with the increased sensitivity of the SERS (Surface Enhanced Raman Spectroscopy) approach and the nanometric spatial resolution of the Scanning Probe Spectroscopy (SPM). Since the technique uses a Raman spectrometer, Tip-Enhanced Photoluminescence (TEPL) is also possible for probing chemical, electrical and optical properties at the nanoscale. The spectroscopy information is provided along with all the measurements enabled by SPM, such as topography and electrical and mechanical properties. The advantages of these spectroscopies, namely the lateral resolution beyond the diffraction limit and the non-destructive, label-free and in-air interaction, open new possibilities for nanotechnology and quantum applications. A new TERS/TEPL equipment was recently installed in the FBK-SD laboratories. This PhD project aims to set up the TERS/TEPL techniques and apply them to the study of optical and morphological properties of semiconductor materials, also nanostructured, used for quantum technologies, down to the single quantum object scale. This research activity will be carried out in the framework of some ongoing projects on quantum sensors and devices at FBK-SD.

The PhD candidate will focus on the development of superconducting quantum devices, such as superconducting qubits and parametric amplifiers, and on the optimisation of the circuit components, such as Josephson junctions and high-kinetic-inductance components. The candidate will be involved in the design and simulations as well as in the microfabrication and in the cryogenic characterisation of the devices. The produced devices will be exploited to investigate fundamental physics cQED processes in the microwave domain and to be integrated in complete quantum systems.

In the last few years, the use of Deep Learning has proved to be very effective in many fields including Bioinformatics and, especially, the study of tumor DNA sequencing data, which allows to measure somatic mutations that accumulate the lifetime of an individual

and are fundamental drivers of cancer. With the final aim of understanding and modeling tumor evolution, in general it is possible to exploit Deep Learning techniques

to extract rich and useful information from datasets of somatic mutations from several cancer patients, which can then be exploited for developing targeted drugs and

therapies for patients. Recent works exploit Deep Reinforcement Learning (DRL) to generate phylogenies belonging to a given cancer cohort. Other works attempt to use DRL to study cancer clones and mutations, opening new scenarios for future improvements.

Graph Neural Networks (GNNs) represent another powerful Deep Learning technique that can be applied to tumor evolution. Indeed, the input data usually consists of trees or, more in general graphs. Therefore, the adoption of GNNs can be of crucial interests, for example to extract rich representations for clones or mutations to be then employed for several prediction tasks. Moreover, this kind of networks can also be exploited to access pathways by providing features that also consider the topological relations in the input graphs.

In this project we will develop advance computational methods based on Deep and Reinforcement Learning for modeling and predicting tumor evolution. We will also use

GNNs to extract relevant features from the input data, so to improve the state of the art for complex prediction tasks.

During the doctoral fellowship, the research activity will be pursued along the following two main lines.

Each dataset can be represented as a similarity graph among patients, enabling the exploration of relationships across various biological and clinical contexts. An innovative strategy involves clustering patients not only independently for each data type but also maximizing similarity across different data layers. This approach facilitates the capture of complex interactions among multiple biological and clinical variables, providing a more comprehensive and integrated view of individual variability. Additionally, a focused approach can be adopted by considering specific gene subsets rather than the entire genome, aiming to concentrate on relevant biological processes or specific areas of interest, thus enriching result interpretation and yielding more targeted insights.

A second intriguing approach involves examining transcriptional regulatory networks or covariance networks among transcripts to identify patient-specific networks. This can be achieved, for example, by monitoring how the covariance matrix changes when a patient is removed from the dataset, thus identifying transcriptional variables of greater relevance to the clinical phenotype. We intend to utilize this information, along with clinical data, to develop predictive models of the clinical phenotype. To this end, we will adopt the answerALS dataset (https://www.answerals.org/) as a case study, providing a rich source of omics and clinical data on patients with amyotrophic lateral sclerosis (ALS), enabling a detailed and in-depth analysis of correlations between molecular variables and clinical manifestations of the disease.

Predictive models based on some form of artificial intelligence are nowadays ubiquitous, with applications ranging from biology to engineering. In the analysis of complex systems, a pressing question emerges: is it possible to forego models traditionally derived from first principles in favor of insights gleaned directly from complex data collections? This project aims at developing a groundbreaking approach that merges the best of two worlds: the precision of classical analytical techniques, based on the theory of dynamical systems, with the adaptive power of deep learning methods. By pursuing this synthesis, the project aims to unlock new dimensions in understanding and predicting the behavior of complex dynamical systems. Practical applications will include predictive modeling of systems of epidemiological interests and population dynamics in general.

This scholarship aims to advance a novel analytical framework that bridges the gap between macroscopic, mesoscopic and microscopic environmental influences on population dynamics of localized communities, in agreement with the objectives of the Human Frontiers Science Program co-funding this project.

The applicative side of the project aims to uncover how these microscopic factors modify the transmissibility of viruses, by focusing on the intricate ways in which localized environmental conditions, such as land use changes and the structure of localized communities, affect vector-borne disease transmission. By developing and applying a methodology that integrates multilayer network analysis with detailed environmental and ecological data, this research will provide nuanced insights into the complex interplay between environmental factors and disease spread. The ultimate goal is to enhance predictive models of complex population dynamics by incorporating a deeper understanding of the environmental determinants at play, offering a more granular approach to epidemic preparedness and vector control strategies.

The goal of this scholarship is two-fold. On the one hand, develop a novel methodology to investigate multilayer network effects on infection-like dynamics, competition among species, dispersal and community response in a changing environment, in agreement with the objectives of the Human Frontiers Science Program co-funding this project.

On the other hand, consider the specific application to the influence of the global human mobility network on the spread of vector-borne diseases (with special focus on Dengue, Zika, and Chikungunya viruses) within Italy.

This project aims to offer strategic insights into epidemic preparedness and response strategies in the face of environmental changes and dispersal patterns.

We are seeking a PhD candidate interested in working at the design and experimentation of AI-powered digital interventions for the prevention and management of chronic diseases. The interventions will be based on evidence-based approaches and protocols for the treatment of these conditions. Background expertise on human-centered design methods, psychology and computer science are welcome to inform the design of the digital interventions.

The PhD candidate will both investigate and contribute to advancing the state of the art of gamification approaches in the design of the user engagement with digital therapeutics for mental well-being. The intervention design will be based on evidence-based protocols for mental health and well-being as active ingredients of the treatment, as well as different types of excipients for the intervention (conversational agents, VR environments, serious games, etc.). Background expertise in the areas of HCI, psychology and computer science is welcome.

We seek a candidate for a PhD position in neuro-symbolic AI, with expertise in both neural networks and symbolic reasoning techniques. Neuro-symbolic AI refers to hybrid approaches to artificial intelligence that combine machine learning (and in particular deep neural networks) with symbolic reasoning techniques. These approaches aim to leverage the strengths of both paradigms: neural networks excel at learning patterns and extracting features from large datasets, while symbolic systems can provide abstract interpretable representations and logic-based or algorithmic reasoning. The candidate will have to study and develop neuro-symbolic systems that address some limitations in current AI systems, such as the difficulty in generalizing knowledge across different domains or tasks. This may involve designing novel architectures that integrate neural and symbolic components, developing algorithms for learning and reasoning with hybrid representations, and exploring new applications where neuro-symbolic approaches can offer some advantages over purely neural or symbolic methods.