arXiv subjects

Giuseppe Mangioni

Publications and source records attributed to Giuseppe Mangioni.

18 recordsLinked to original sources

arXiv2026

ProtoGuide: Prototype-Driven Guidance for Class-Conditional Graph Generation

Discrete diffusion models are a prominent family for graph generation, but standard class-conditional mechanisms embed the class signal in the denoiser during training, tying the conditioning mechanism to the trained model. Classifier guidance avoids this coupling in continuous domains by steering a frozen model with a classifier's gradient, but discrete graph diffusion samples discrete edge states, so gradients cannot propagate through the sampled graph. We introduce ProtoGuide, a post-hoc, backbone-agnostic framework that recovers an analogous mechanism. At each reverse step the denoiser's per-edge output is relaxed into a differentiable soft adjacency, embedded by a frozen Siamese graph neural network, and scored against a target-class prototype and its nearest competitor; the resulting per-edge gradient, damped by a cosine schedule, is injected back into the denoiser output. All components stay frozen, so guidance is retargeted by supplying a different prototype. On five classes of real-world networks and two architecturally different backbones, EDGE and DiGress, ProtoGuide raises macro classification accuracy from 50.7% to 73.5% and from 73.6% to 83.8%, and outperforms DiGress's built-in conditional training under our configuration. Gains are largest where the unguided models are weakest, and are not uniform across classes. Per-graph coverage remains high in most settings, while distributional effects are class-dependent. A Best-of-N selection baseline matches this accuracy given enough oversampling, but at a substantial cost in graph diversity. An independently initialized classifier, a directionality test, and a few-shot analysis support target-directed steering and robustness to very small support sets.

cs.LG↗

arXiv2026

Brain-Grasp: Graph-based Saliency Priors for Improved fMRI-based Visual Brain Decoding

Recent progress in brain-guided image generation has improved the quality of fMRI-based reconstructions; however, fundamental challenges remain in preserving object-level structure and semantic fidelity. Many existing approaches overlook the spatial arrangement of salient objects, leading to conceptually inconsistent outputs. We propose a saliency-driven decoding framework that employs graph-informed saliency priors to translate structural cues from brain signals into spatial masks. These masks, together with semantic information extracted from embeddings, condition a diffusion model to guide image regeneration, helping preserve object conformity while maintaining natural scene composition. In contrast to pipelines that invoke multiple diffusion stages, our approach relies on a single frozen model, offering a more lightweight yet effective design. Experiments show that this strategy improves both conceptual alignment and structural similarity to the original stimuli, while also introducing a new direction for efficient, interpretable, and structurally grounded brain decoding.

eess.IV↗

arXiv2026

Beyond MMD: Evaluating Graph Generative Models with Geometric Deep Learning

Graph generation is a crucial task in many fields, including network science and bioinformatics, as it enables the creation of synthetic graphs that mimic the properties of real-world networks for various applications. Graph Generative Models (GGMs) have emerged as a promising solution to this problem, leveraging deep learning techniques to learn the underlying distribution of real-world graphs and generate new samples that closely resemble them. Examples include approaches based on Variational Auto-Encoders, Recurrent Neural Networks, and more recently, diffusion-based models. However, the main limitation often lies in the evaluation process, which typically relies on Maximum Mean Discrepancy (MMD) as a metric to assess the distribution of graph properties in the generated ensemble. This paper introduces a novel methodology for evaluating GGMs that overcomes the limitations of MMD, which we call RGM (Representation-aware Graph-generation Model evaluation). As a practical demonstration of our methodology, we present a comprehensive evaluation of two state-of-the-art Graph Generative Models: Graph Recurrent Attention Networks (GRAN) and Efficient and Degree-guided graph GEnerative model (EDGE). We investigate their performance in generating realistic graphs and compare them using a Geometric Deep Learning model trained on a custom dataset of synthetic and real-world graphs, specifically designed for graph classification tasks. Our findings reveal that while both models can generate graphs with certain topological properties, they exhibit significant limitations in preserving the structural characteristics that distinguish different graph domains. We also highlight the inadequacy of Maximum Mean Discrepancy as an evaluation metric for GGMs and suggest alternative approaches for future research.

cs.LG↗

arXiv2025

Robustness and resilience of complex networks

Complex networks are ubiquitous: a cell, the human brain, a group of people and the Internet are all examples of interconnected many-body systems characterized by macroscopic properties that cannot be trivially deduced from those of their microscopic constituents. Such systems are exposed to both internal, localized, failures and external disturbances or perturbations. Owing to their interconnected structure, complex systems might be severely degraded, to the point of disintegration or systemic dysfunction. Examples include cascading failures, triggered by an initially localized overload in power systems, and the critical slowing downs of ecosystems which can be driven towards extinction. In recent years, this general phenomenon has been investigated by framing localized and systemic failures in terms of perturbations that can alter the function of a system. We capitalize on this mathematical framework to review theoretical and computational approaches to characterize robustness and resilience of complex networks. We discuss recent approaches to mitigate the impact of perturbations in terms of designing robustness, identifying early-warning signals and adapting responses. In terms of applications, we compare the performance of the state-of-the-art dismantling techniques, highlighting their optimal range of applicability for practical problems, and provide a repository with ready-to-use scripts, a much-needed tool set.

physics.soc-ph↗

arXiv2021

The rise and fall of countries in the global value chains

Countries participate in global value chains by engaging in backward and forward transactions connecting multiple geographically dispersed production stages. Inspired by network theory, we model global trade as a multilayer network and study its power structure by investigating the tendency of eigenvector centrality to concentrate on a small fraction of countries, a phenomenon called localization transition. We show that the market underwent a significant structural variation in 2007 just before the global financial crisis. That year witnessed an abrupt repositioning of countries in the global value chains, and in particular a remarkable reversal of leading role between the two major economies, the US and China. We uncover the hierarchical structure of the multilayer network based on countries' time series of eigenvector centralities, and show that trade tends to concentrate between countries with different power dynamics, yet in geographical proximity. We further investigate the contribution of individual industries to countries' economic dominance, and show that also within industry variations in countries' market positioning took place in 2007. Moreover, we shed light on the crucial role that domestic trade played in the geopolitical landscape leading China to overtake the US and cement its status as leading economy of the global value chains. Our study shows how the 2008 crisis can offer insights to policy makers and governments on how to turn early structural signals of upcoming exogenous shocks into opportunities for redesigning countries' global roles in a changing geopolitical landscape.

physics.soc-ph↗

arXiv2021

mGNN: Generalizing the Graph Neural Networks to the Multilayer Case

Networks are a powerful tool to model complex systems, and the definition of many Graph Neural Networks (GNN), Deep Learning algorithms that can handle networks, has opened a new way to approach many real-world problems that would be hardly or even untractable. In this paper, we propose mGNN, a framework meant to generalize GNNs to the case of multi-layer networks, i.e., networks that can model multiple kinds of interactions and relations between nodes. Our approach is general (i.e., not task specific) and has the advantage of extending any type of GNN without any computational overhead. We test the framework into three different tasks (node and network classification, link prediction) to validate it.

cs.LG↗

arXiv2021

wsGAT: Weighted and Signed Graph Attention Networks for Link Prediction

Graph Neural Networks (GNNs) have been widely used to learn representations on graphs and tackle many real-world problems from a wide range of domains. In this paper we propose wsGAT, an extension of the Graph Attention Network (GAT) layers, meant to address the lack of GNNs that can handle graphs with signed and weighted links, which are ubiquitous, for instance, in trust and correlation networks. We first evaluate the performance of our proposal by comparing against GCNII in the weighed link prediction task, and against SGCN in the link sign prediction task. After that, we combine the two tasks and show their performance on predicting the signed weight of links, and their existence. Our results on real-world networks show that models with wsGAT layers outperform the ones with GCNII and SGCN layers, and that there is no loss in performance when signed weights are predicted.

cs.SI↗

arXiv2021

Machine learning dismantling and early-warning signals of disintegration in complex systems

From physics to engineering, biology and social science, natural and artificial systems are characterized by interconnected topologies whose features - e.g., heterogeneous connectivity, mesoscale organization, hierarchy - affect their robustness to external perturbations, such as targeted attacks to their units. Identifying the minimal set of units to attack to disintegrate a complex network, i.e. network dismantling, is a computationally challenging (NP-hard) problem which is usually attacked with heuristics. Here, we show that a machine trained to dismantle relatively small systems is able to identify higher-order topological patterns, allowing to disintegrate large-scale social, infrastructural and technological networks more efficiently than human-based heuristics. Remarkably, the machine assesses the probability that next attacks will disintegrate the system, providing a quantitative method to quantify systemic risk and detect early-warning signals of system's collapse. This demonstrates that machine-assisted analysis can be effectively used for policy and decision making to better quantify the fragility of complex systems and their response to shocks.

physics.soc-ph↗

arXiv2020

(Unintended) Consequences of export restrictions on medical goods during the Covid-19 pandemic

In the first half of 2020, several countries have responded to the challenges posed by the Covid-19 pandemic by restricting their export of medical supplies. Such measures are meant to increase the domestic availability of critical goods, and are commonly used in times of crisis. Yet, not much is known about their impact, especially on countries imposing them. Here we show that export bans are, by and large, counterproductive. Using a model of shock diffusion through the network of international trade, we simulate the impact of restrictions under different scenarios. We observe that while they would be beneficial to a country implementing them in isolation, their generalized use makes most countries worse off relative to a no-ban scenario. As a corollary, we estimate that prices increase in many countries imposing the restrictions. We also find that the cost of restraining from export bans is small, even when others continue to implement them. Finally, we document a change in countries' position within the international trade network, suggesting that export bans have geopolitical implications.

physics.soc-ph↗

arXiv2019

The nested structural organization of the worldwide trade multi-layer network

Nestedness has traditionally been used to detect assembly patterns in meta-communities and networks of interacting species. Attempts have also been made to uncover nested structures in international trade, typically represented as bipartite networks in which connections can be established between countries (exporters or importers) and industries. A bipartite representation of trade, however, inevitably neglects transactions between industries. To fully capture the organization of the global value chain, we draw on the World Input-Output Database and construct a multi-layer network in which the nodes are the countries, the layers are the industries, and links can be established from sellers to buyers within and across industries. We define the buyers' and sellers' participation matrices in which the rows are the countries and the columns are all possible pairs of industries, and then compute nestedness based on buyers' and sellers' involvement in transactions between and within industries. Drawing on appropriate null models that preserve the countries' or layers' degree distributions in the original multi-layer network, we uncover variations of country- and transaction-based nestedness over time, and identify the countries and industries that most contributed to nestedness. We discuss the implications of our findings for the study of the international production network and other real-world systems.

physics.soc-ph↗

arXiv2019

Modularity in Multilayer Networks using Redundancy-based Resolution and Projection-based Inter-Layer Coupling

The generalized version of modularity for multilayer networks, a.k.a. multislice modularity, is characterized by two model parameters, namely resolution factor and inter-layer coupling factor. The former corresponds to a notion of layer-specific relevance, whereas the inter-layer coupling factor represents the strength of node connections across the network layers. Despite the potential of this approach, the setting of both parameters can be arbitrarily selected, without considering specific characteristics from the topology of the multilayer network as well as from an available community structure. Also, the multislice modularity is not designed to explicitly model order relations over the layers, which is of prior importance for dynamic networks. This paper aims to overcome the main limitations of the multislice modularity by introducing a new formulation of modularity for multilayer networks. We revise the role and semantics of both the resolution and inter-layer coupling factors based on information available from the within-layer and inter-layer structures of the multilayer communities. Also, our proposed multilayer modularity is general enough to consider orderings of network layers and their constraints on layer coupling. Experiments were carried out on synthetic and real-world multilayer networks using state-of-the-art approaches for multilayer community detection. The obtained results have shown the meaningfulness of the proposed modularity, revealing the effects of different combinations of the resolution and inter-layer coupling functions. This work also represents a starting point for the development of new optimization methods for community detection in multilayer networks.

cs.SI↗

arXiv2018

Unfolding the complexity of the global value chain: Strengths and entropy in the single-layer, multiplex, and multi-layer international trade networks

The worldwide trade network has been widely studied through different data sets and network representations with a view to better understanding interactions among countries and products. Here we investigate international trade through the lenses of the single-layer, multiplex, and multi-layer networks. We discuss differences among the three network frameworks in terms of their relative advantages in capturing salient topological features of trade. We draw on the World Input-Output Database to build the three networks. We then uncover sources of heterogeneity in the way strength is allocated among countries and transactions by computing the strength distribution and entropy in each network. Additionally, we trace how entropy evolved, and show how the observed peaks can be associated with the onset of the global economic downturn. Findings suggest how more complex representations of trade, such as the multi-layer network, enable us to disambiguate the distinct roles of intra- and cross-industry transactions in driving the evolution of entropy at a more aggregate level. We discuss our results and the implications of our comparative analysis of networks for research on international trade and other empirical domains across the natural and social sciences.

physics.soc-ph↗

arXiv2018

Black Hole Metric: Overcoming the PageRank Normalization Problem

In network science, there is often the need to sort the graph nodes. While the sorting strategy may be different, in general sorting is performed by exploiting the network structure. In particular, the metric PageRank has been used in the past decade in different ways to produce a ranking based on how many neighbors point to a specific node. PageRank is simple, easy to compute and effective in many applications, however it comes with a price: as PageRank is an application of the random walker, the arc weights need to be normalized. This normalization, while necessary, introduces a series of unwanted side-effects. In this paper, we propose a generalization of PageRank named Black Hole Metric which mitigates the problem. We devise a scenario in which the side-effects are particularily impactful on the ranking, test the new metric in both real and synthetic networks, and show the results.

cs.SI↗

arXiv2018

Multilayer flows in molecular networks identify biological modules in the human proteome

A variety of complex systems exhibit different types of relationships simultaneously that can be modeled by multiplex networks. A typical problem is to determine the community structure of such systems that, in general, depend on one or more parameters to be tuned. In this study we propose one measure, grounded on information theory, to find the optimal value of the relax rate characterizing Multiplex Infomap, the generalization of the Infomap algorithm to the realm of multilayer networks. We evaluate our methodology on synthetic networks, to show that the most representative community structure can be reliably identified when the most appropriate relax rate is used. Capitalizing on these results, we use this measure to identify the most reliable meso-scale functional organization in the human protein-protein interaction multiplex network and compare the observed clusters against a collection of independently annotated gene sets from the Molecular Signatures Database (MSigDB). Our analysis reveals that modules obtained with the optimal value of the relax rate are biologically significant and, remarkably, with higher functional content than the ones obtained from the aggregate representation of the human proteome. Our framework allows us to characterize the meso-scale structure of those multilayer systems whose layers are not explicitly interconnected each other -- as in the case of edge-colored models -- the ones describing most biological networks, from proteomes to connectomes.

q-bio.MN↗

arXiv2017

Identifying the community structure of the international food-trade multi network

Achieving international food security requires improved understanding of how international trade networks connect countries around the world through the import-export flows of food commodities. The properties of food trade networks are still poorly documented, especially from a multi-network perspective. In particular, nothing is known about the community structure of food networks, which is key to understanding how major disruptions or 'shocks' would impact the global food system. Here we find that the individual layers of this network have densely connected trading groups, a consistent characteristic over the period 2001 to 2011. We also fit econometric models to identify social, economic and geographic factors explaining the probability that any two countries are co-present in the same community. Our estimates indicate that the probability of country pairs belonging to the same food trade community depends more on geopolitical and economic factors -- such as geographical proximity and trade agreements co-membership -- than on country economic size and/or income. This is in sharp contrast with what we know about bilateral-trade determinants and suggests that food country communities behave in ways that can be very different from their non-food counterparts.

q-fin.GN↗

arXiv2016

Communities unfolding in multislice networks

Discovering communities in complex networks helps to understand the behaviour of the network. Some works in this promising research area exist, but communities uncovering in time-dependent and/or multiplex networks has not deeply investigated yet. In this paper, we propose a communities detection approach for multislice networks based on modularity optimization. We first present a method to reduce the network size that still preserves modularity. Then we introduce an algorithm that approximates modularity optimization (as usually adopted) for multislice networks, thus finding communities. The network size reduction allows us to maintain acceptable performances without affecting the effectiveness of the proposed approach.

physics.soc-ph↗

arXiv2010

Identifying the Community Structure of the International-Trade Multi Network

We study the community structure of the multi-network of commodity-specific trade relations among world countries over the 1992-2003 period. We compare structures across commodities and time by means of the normalized mutual information index (NMI). We also compare them with exogenous community structures induced by geographical distances and regional trade agreements. We find that commodity-specific community structures are very heterogeneous and much more fragmented than that characterizing the aggregate ITN. This shows that the aggregate properties of the ITN may result (and be very different) from the aggregation of very diverse commodity-specific layers of the multi network. We also show that commodity-specific community structures, especially those related to the chemical sector, are becoming more and more similar to the aggregate one. Finally, our findings suggest that geographical distance is much more correlated with the observed community structure than RTAs. This result strengthens previous findings from the empirical literature on trade.

physics.soc-ph↗

arXiv2009

Search for overlapped communities by parallel genetic algorithms

In the last decade the broad scope of complex networks has led to a rapid progress. In this area a particular interest has the study of community structures. The analysis of this type of structure requires the formalization of the intuitive concept of community and the definition of indices of goodness for the obtained results. A lot of algorithms has been presented to reach this goal. In particular, an interesting problem is the search of overlapped communities and it is field seems very interesting a solution based on the use of genetic algorithms. The approach discusses in this paper is based on a parallel implementation of a genetic algorithm and shows the performance benefits of this solution.

cs.IR↗