Higher Institute on Territorial Systems for Innovation
PORTO@iris (Publications Open Repository TOrino - Politecnico di Torino)Not a member yet
146173 research outputs found
Sort by
Is it possible a `Reverse Flynn Effect'? Consideration for new opportunities in the context of sustainability
No full textRecently, some evaluations of the workers’ and students’ abilities brought
back to consider the negative (or reverse) Flynn effect. The reverse Flynn effect is an
empirically confirmed decrease in the Intelligence Quotient since the 1990s. It has been
proven to be related to environmental conditions only. Here, we consider this effect as an
environmental consequence of the social and education conditions, considering it a possible
resource, meaning the opportunity to reconsider the education pathway and the workers’
training. To do so, in this paper, we develop a thermodynamic analysis of the information
fluxes that enter the brain and the related stabilisation of the inflow of information itself.
Moreover, we develop a thermo-economic analysis of the consequences of the reverse Flynn
effect, pointing out the need to focus educational policies on the continuous stimulus of the
use of reasoning and problem-solving-based education, to develop the processing capacity
and foster the creative capabilities of young people, who build the backbone of the future
workforce
Measuring Firm Greenness: A Comprehensive Review of Corporate Sustainability Indicators
No full textThe major global challenges of climate change, environmental degradation, and resource scarcity have gained centrality in
policymaking and academia, with a growing body of research exploring the interlinkages among economic growth, trade,
environmental outcomes, and policy interventions. These analyses and theirresults draw fundamentally on the definitions used to
consider a firm, product, or process “green,” and on the related operationalization. Despite a proliferation of empirical approaches,
however, the literature lacks a comprehensive overview of firm-level sustainability indicators. This article addresses this gap by
systematically reviewing existing definitions and the most frequently employed indicators Our analysis shows that, according
to prevailing definitions of greenness, a firm can be considered green either because it is purpose-sustainable, meaning that it
primarily serves an environmental purpose, or because it is process-sustainable, meaning that its activities reduce or prevent
negative environmental impacts even if environmental goals are not its primary purpose. We further classify the existing firmlevel sustainability indicators into three groups: (1) product-level indicators; (2) resource and pollution management indicators;
and (3) investment, innovation, and commitment indicators. Our findings aim to support researchers in selecting indicators that
align with specific research objectives and dimensions of sustainability under investigatio
Investigation of selected invasion problems using analytical, numerical, and control approaches
No full textL'abstract è presente nell'allegato / the abstract is in the attachmen
Advanced multiscale models for the analysis of composite structures with applications to curing simulation and multiphysics
No full textL'abstract è presente nell'allegato / the abstract is in the attachmen
A Systematic Literature Review on Disruptions in Construction Supply Chain: Some Stylized Trends
No full textConstruction supply chains (CSC) are intricate systems characterized by fragmentation and inefficiencies, which have been exacerbated by external disruptions such as the COVID-19 pandemic, natural disasters, and geopolitical instabilities. These challenges underscore the urgent need for resilient and adaptive supply chain strategies that integrate emerging technologies. However, existing research remains fragmented, lacking a comprehensive framework that aligns digital innovations with broader operational, economic, and stakeholder considerations. To bridge this gap, this study employs a systematic literature review (SLR) complemented by natural language processing (NLP) techniques to analyze existing knowledge on CSC disruptions and resilience-building strategies. A total of 63 articles were reviewed, with thematic clustering and topic modeling applied to uncover key challenges and opportunities. Findings reveal two primary clusters: macro-level strategies focused on systemic resilience and micro-level solutions addressing operational inefficiencies. The results indicate that while Industry 4.0 and modular construction methods offer promising solutions, their integration into CSC frameworks remains inconsistent. By combining qualitative SLR insights with quantitative NLP analysis, this study provides a holistic perspective on CSC disruptions and potential resilience strategies, offering valuable implications for both academia and industry. In particular, it might support practitioners in identifying suitable solutions for material tracking and in turn reducing inefficiencies. In addition, the issue of collaboration and information sharing is stressed in order to achieve a more aware decision-making process and reduce the level of uncertainty with a consequent higher resilience along the supply chain in the construction industry
Enhancing Manufacturing Engineering Higher Education Through Mixed Reality and Gaussian Splatting: Preliminary Experimental Results
No full textThe integration of Mixed Reality (MR) technologies into higher manufacturing-engineering education can contribute to face the challenges of providing hands-on training with real manufacturing systems. This paper explores the potential of MR combined with Gaussian Splatting (GS) to create high-fidelity digital replicas of industrial machinery (e.g., lathes, milling machines, etc.), enhancing students’ understanding of manufacturing processes. GS is emerging as a breakthrough technique for real-time rendering of objects and environments. By delineating the scene as the realisation of an object in a defined temporal state, GS methodology represents a 3D high-fidelity digital scene as a collection of 3D Gaussian ellipsoids characterised by position, geometry, shape, colour and opacity. The integration of MR with GS allows trainees to engage with realistic virtual models, simulating a physical presence in a machining workshop. The capacity to digitally manipulate and analyse individual objects enhances the learning experience, addressing logistical and safety constraints by providing a risk-free and accessible training environment. A lathe is used as a case study, and the GS-based digital scene is compared with conventional CAD-based model in terms of qualitative performance
Boosting zero-shot learning through neuro-symbolic integration
Get PDFZero-shot learning (ZSL) aims to train deep neural networks to recognize objects from unseen classes, starting from a semantic description of the concepts. Neuro-symbolic (NeSy) integration refers to a class of techniques that incorporate symbolic knowledge representation and reasoning with the learning capabilities of deep neural networks. However, to date, few studies have explored how to leverage NeSy techniques to inject prior knowledge during the training process to boost ZSL capabilities. Here, we present Fuzzy Logic Prototypical Network (FLPN) that formulates the classification task as prototype matching in a visual-semantic embedding space, which is trained by optimizing a NeSy loss. Specifically, FLPN exploits the Logic Tensor Network (LTN) framework to incorporate background knowledge in the form of logical axioms by grounding a first-order logic language as differentiable operations between real tensors. This prior knowledge includes class hierarchies (classes and macroclasses) along with robust high-level inductive biases. The latter allow, for instance, to handle exceptions in class-level attributes and to enforce similarity between images of the same class, preventing premature overfitting to seen classes and improving overall performance. Both class-level and attribute-level prototypes through an attention mechanism specialized for either convolutional- or transformer-based backbones. FLPN achieves state-of-the-art performance on the GZSL benchmarks AWA2 and SUN, matching or exceeding the performance of competing algorithms with minimal computational overhead.
The code is available at https://github.com/FrancescoManigrass/FLP
Boosting biocompatibility and minimizing inflammation in electrospun polyvinylidene fluoride (PVDF) cardiac patches through optimized low-pressure plasma treatment
Get PDF: Tailoring surface characteristics is key to guiding scaffold interaction with the biological environment, promoting successful biointegration while minimizing immune responses and inflammation. In cardiac tissue engineering, polyvinylidene fluoride (PVDF) is a material of choice for its intrinsic piezoelectric properties, which can be enhanced through electrospinning, also enabling the fabrication of nanofibrous structures mimicking native tissue. However, the inherent hydrophobicity of PVDF can hinder its integration with biological tissues. To overcome this limitation, electrospun PVDF patches were subjected to radio-frequency low-pressure O2 plasma treatment to enhance surface hydrophilicity and overall biocompatibility. A systematic experimental study identified optimal parameters, revealing that higher gas content and prolonged exposure are preferable to high power levels, which deteriorate the patch's morphological and mechanical properties. X-ray photoelectron spectroscopy confirmed the formation of oxygen-containing surface groups, resulting in the patch's superhydrophilicity. Preservation of the fibrous nanostructure and electroactive phase content was verified using scanning electron microscopy and infrared spectroscopy combined with differential scanning calorimetry, respectively. The optimized plasma treatment maintained the patch's elasticity and demonstrated long-term stability for up to 3 months. In vitro biocompatibility was assessed through indirect and direct tests using AC16 human cardiomyocytes and neonatal human dermal fibroblasts, revealing good cell viability, adhesion, and spreading over 7-days. Finally, plasma-treated patches demonstrated strong adhesion to the myocardial tissue and exhibited markedly reduced inflammatory response compared to the untreated controls, as shown by decreased CD45+ immune cell infiltration around the patch implanted in infarcted mice, highlighting the surface treatment's effectiveness in enhancing in vivo biocompatibility
Refined Two-Sided Learning Rate Tuning for Robust Evaluation in Federated Learning
No full textThis paper investigates the impact of client and server learning rates on training deep neural networks in Federated Learning (FL). While previous research has primarily focused on optimizing the initial values of these learning rates, we demonstrate that this approach alone is insufficient for maximizing model performance and training efficiency. To address this weakness, we propose a revised two-sided learning rate optimization strategy that integrates learning rate decay schedules as tunable variables and adjusts the learning rate configurations based on the target training budget, allowing for more effective optimization.
We conduct an extensive experimental evaluation to quantify the improvements offered by our approach. The results reveal that (i) integrating decay schedules into the tuning process leads to significant performance enhancements, and (ii) the optimal configuration of client-server decay schedules is strongly influenced by the training round budget. Based on these findings, we claim that performance evaluations of new FL algorithms should extend beyond the fine-tuning of the initial learning rate values, as done in the state-of-the-art approach, and include the optimization of decay schedules according to the available training budget
Quantum open system description of a hybrid plasmonic cavity
No full textWe present a unified quantum open system framework for lossy plasmonic cavities, treating coherent dynamics, relaxation, dephasing, and irreversible absorption on an equal footing. The Dyson equation for the cavity photon propagator in the random-phase approximation yields a complex self-energy S(ω) that accounts for both the renormalization and damping of hybrid plasmon-photon modes. It shows that increasing losses can drive a crossover from resolvable normal-mode splitting to a regime without resolved splitting, when the damping becomes comparable to or larger than the coherent hybridization scale. Tracing out the environment yields a Liouvillian for the upper polaritons (UPs) and lower polaritons (LPs) with leakage \Gamma = −2 ImS(ω), internal UP ↔ LP scattering, and dephasing. Closed-form dynamics for populations and interbranch coherence provide analytic steady-state values, line shapes, and UP-LP quench rates, valid at low polariton density and in the ultrastrong-coupling regime. The theory is directly applicable to spectra, time-domain probes, and dissipation engineering in plasmonic and nanophotonic cavities