The exploration of topological singularities is a fascinating task of absolute relevance both from the theoretical and applied point of view. For example, in physics and materials science they arise from the study of mathematical models for vortices in superconductors, grain boundaries in polycrystals, fractures in solids, and defects in crystals such as disclinations or dislocations. Furthermore, topological singularities play an important role in the study of more geometric problems such as the Plateau problem and the theory of minimal surfaces. The goal of TopSing is to study some physical/mechanical problems where singularities appear, through a theoretical approach that opens promising directions also for other, apparently unrelated, situations like the non-parametric Plateau problem in higher codimension. More specifically, we draw our attention to codimension-two singularities and consider two-dimensional models for fields having point singularities which are relevant in the study of two main problems: 1) Screw Dislocations in crystals and their relation with vortices in superconductors; 2) The non-parametric Plateau problem in codimension-two. The main novelty consists in developing a unified approach, inspired by the classical model by Ambrosio and Tortorelli (AT), that allows to study topological singularities in both contexts listed above. Furthermore this will provide a model which is easier to handle numerically and thus interesting from the point of view of applications. The project is organised into four main objectives whose common thread is the asymptotic analysis of elliptic functionals á la AT for maps taking values on the unit circle. To our best knowledge there are no similar results in the literature. This is due to the non trivial task of constructing a recovery sequence that takes values on the circle, which we aim at achieving by relying on degree theory and by using techniques developed to study the relaxed area.

EcoAI aims at creating a new regulatory framework for an ecological AI to align AI development with sustainability, bridging gaps in scholarly efforts but also providing policy suggestions to EU institutions, facilitating evidence-based decision making, and ensuring that the proposed regulatory framework is a blueprint both for technology regulation more broadly, and for non-EU Countries (Brazil will be a case study in this respect). Other Countries’ legislation could also be influenced by signing cooperation agreement on the sustainable use of AI. It is for this reason that EcoAI aims also at drafting a proposal for an agreement between the EU and Brazil to strengthen their cooperation in addressing the global challenge of climate change using AI. To build this new framework and the proposal for an agreement on AI for the public good between the EU and Brazil, I will develop and test in intersectoral practical environments and in two different legal frameworks, i.e. EU and Brazil, the innovative concept of “Ecological AI”: the possibility of using AI in an “ecological” way for the public good, in accordance with the fundamental rights enshrined in the Charter of Fundamental Rights of the EU . To reach this objective, EcoAI has a strong interdisciplinary approach, bridging law, computer science, and political sciences, with some exposure to sociology, taking into primary account the needs of the public and private sector, and those of citizens, who are the category mainly exposed to the negative effects of climate change. To ensure the successful implementation of these project goals, to gain new skills and competences, and to identify solutions to two of the biggest challenges of our time, the University of Siena (UNISI) and the University of San Paolo (USP) are the most appropriate venues. This research directly contribute to achieving Horizon Europe, the European Green Deal, the European Data Strategy, the Digital Compass 2030 Targets, and the SDGs.

Narratives – the stories people formulate to make sense of the world – contain at their core cause and effect relationships that people use to connect the events they observe. These relationships are crucial for interpreting data, forming expectations, and evaluating policy proposals. This project aims at improving our understanding of the role of economic narratives, their interaction with policy proposals, and how they can be manipulated by economic actors. The ultimate goal of this project is to suggest strategies to follow when faced with contexts in which struggles over policies might generate polarized views of how the world works. The project is divided into three areas and combines ideas and methods at the intersection between economics, sociology, psychology, and computer science. The first area of research involves behavioral experiments in which I will study how narratives react to the cost of policies available to agents. In particular, I will investigate whether agents tend to adopt narratives that are self-serving, that is, compatible with the policies they find least costly. The second area of research focuses on the role that narratives played in shaping political preference in the second half of the XX century. I will consider labor unions to study the factors behind the emergence in the United States of a narrative highlighting their mischiefs. These results could highlight the magnitude and importance of the consequences generated by those that become prevailing narratives in a society. The last area of research will involve a methodological study on how to measure narratives from text. I will work to extend the available methods with the goal of explicitly capturing causal relationships. This contribution could favor the emergence of new work in the field by other researchers who could easily analyze large corpora of text.

The antimicrobial resistance (AMR) is an underrated problem promoted by widespread misuse and massive use of antibiotics. Thus, the health systems are facing enormous costs to treat nosocomial infections due to AMR bacteria. In particular, six highly virulent bacteria (E.S.K.A.P.E.) are the main responsible for worldwide nosocomial infections. In the last years, our research strategies to discover and develop new clinically relevant antibiotics lacked effectiveness. As a result, we are unable to efficiently treat AMR infections and are unprepared to face an AMR bacteria outbreak. One possible strategy is to target bacteria G-quadruplexes (bGQs) which are highly conserved DNA/RNA secondary structures pivotal for bacteria duplication, transcription and translation. This strategy is particular effective because, when a new pathogen emerges, its DNA/RNA genome is first disclosed through sequencing, hence the ability of quickly designing new drug candidates based only on this information will be a powerful tool to face a new epidemic/pandemic. Furthermore, the human GQs (hGQs) were largely investigated as possible target for anticancer treatment thus the interest in bGQs raised later and with less intensity than hGQs. In addition, most GQs modulators bind G-quartets top by π-stacking interactions and the druggability of GQs loops/grooves was still not addressed. In this context, the multidisciplinary of G-Q-reat ESKAPE project aims to provide a novel and effective strategy to contrast AMR by targeting bGQs and to expand the knowledge on bGQs.

The original HARIA project aims to build human sensorimotor augmentation systems, particularly for people with uni- or bi-lateral upper-limb chronic motor disabilities. This new approach to physical human-robot interaction is achieved through the development of supernumerary robotic limbs and wearable sensorimotor interfaces. The combination of these technologies, together with Artificial Intelligence methodologies, allows the users to control and feel the robotic limbs as an extension of their bodies. In this Hop-On Facility activity, we propose to expand the ambition of the HARIA project, by adding a robust layer that manages the occurrence of failures and anomalous situations, using a combination of methods to Predict, Detect, Communicate, and Recover from failure events during the execution of collaborative and augmented tasks, aimed at the application scenarios envisaged in HARIA. Prediction of failure events will be done through a Deep Learning classifier using multi-modal sensory information, while actual failure detection relies mostly on model-based methods. Failures are communicated to the users via the Wearable Sensorimotor Interfaces developed in the project, who can then control the reaction strategies. HARIA-FT strives to improve the robustness of the original system through this failure management framework and through extensive experimentation and validation.