Search

Signed graphs allow for encoding positive and negative relations between nodes and are used to model various online activities. Node representation learning for signed graphs is a well-studied task with important applications such as sign prediction. While the size of datasets is ever-increasing, recent methods often sacrifice scalability for accuracy. We propose a novel message-passing layer architecture called Graph Spring Network (GSN) modeled after spring forces. We combine it with a Graph Neural Ordinary Differential Equations (ODEs) formalism to optimize the system dynamics in embedding space to solve a downstream prediction task. Once the dynamics is learned, embedding generation for novel datasets is done by solving the ODEs in time using a numerical integration scheme. Our GSN layer leverages the fast-to-compute edge vector directions and learnable scalar functions that only depend on nodes' distances in latent space to compute the nodes' positions. Conversely, Graph Convolution and Graph Attention Network layers rely on learnable vector functions that require the full positions of input nodes in latent space. We propose a specific implementation called Spring-Neural-Network (SPR-NN) using a set of small neural networks mimicking attracting and repulsing spring forces that we train for link sign prediction. Experiments show that our method achieves accuracy close to the state-of-the-art methods with node generation time speedup factors of up to 28,000 on large graphs.

Decentralized Federated Learning (DFL) trains models in a collaborative and privacy-preserving manner while removing model centralization risks and improving communication bottlenecks. However, DFL faces challenges in efficient communication management and model aggregation within decentralized environments, especially with heterogeneous data distributions. Thus, this paper introduces ProFe, a novel communication optimization algorithm for DFL that combines knowledge distillation, prototype learning, and quantization techniques. ProFe utilizes knowledge from large local models to train smaller ones for aggregation, incorporates prototypes to better learn unseen classes, and applies quantization to reduce data transmitted during communication rounds. The performance of ProFe has been validated and compared to the literature by using benchmark datasets like MNIST, CIFAR10, and CIFAR100. Results showed that the proposed algorithm reduces communication costs by up to ~40-50% while maintaining or improving model performance. In addition, it adds ~20% training time due to increased complexity, generating a trade-off.

Identifying significant community structures in networks with incomplete data is a challenging task, as the reliability of solutions diminishes with increasing levels of missing information. However, in many empirical contexts, some information about the uncertainty in the network measurements can be estimated. In this work, we extend the recently developed Flow Stability framework, originally designed for detecting communities in time-varying networks, to address the problem of community detection in weighted, directed networks with missing links. Our approach leverages known uncertainty levels in nodes' out-degrees to enhance the robustness of community detection. Through comparisons on synthetic networks and a real-world network of messaging channels on the Telegram platform, we demonstrate that our method delivers more reliable community structures, even when a significant portion of data is missing.

Federated Learning (FL) performance is highly influenced by data distribution across clients, and non-Independent and Identically Distributed (non-IID) leads to a slower convergence of the global model and a decrease in model effectiveness. The existing algorithms for solving the non-IID problem are focused on the traditional centralized FL (CFL), where a central server is used for model aggregation. However, in decentralized FL (DFL), nodes lack the overall vision of the federation. To address the non-IID problem in DFL, this paper proposes a novel DFL aggregation algorithm, Federated Entropy Pooling (FedEP). FedEP mitigates the client drift problem by incorporating the statistical characteristics of local distributions instead of any actual data. Prior to training, each client conducts a local distribution fitting using a Gaussian Mixture Model (GMM) and shares the resulting statistical characteristics with its neighbors. After receiving the statistical characteristics shared by its neighbors, each node tries to fit the global data distribution. In the aggregation phase, each node calculates the Kullback-Leibler (KL) divergences of the local data distribution over the fitted global data distribution, giving the weights to generate the aggregated model. Extensive experiments have demonstrated that FedEP can achieve faster convergence and show higher test performance than state-of-the-art approaches.

Federated Learning (FL), introduced in 2016, was designed to enhance data privacy in collaborative model training environments. Among the FL paradigm, horizontal FL, where clients share the same set of features but different data samples, has been extensively studied in both centralized and decentralized settings. In contrast, Vertical Federated Learning (VFL), which is crucial in real-world decentralized scenarios where clients possess different, yet sensitive, data about the same entity, remains underexplored. Thus, this work introduces De-VertiFL, a novel solution for training models in a decentralized VFL setting. De-VertiFL contributes by introducing a new network architecture distribution, an innovative knowledge exchange scheme, and a distributed federated training process. Specifically, De-VertiFL enables the sharing of hidden layer outputs among federation clients, allowing participants to benefit from intermediate computations, thereby improving learning efficiency. De-VertiFL has been evaluated using a variety of well-known datasets, including both image and tabular data, across binary and multiclass classification tasks. The results demonstrate that De-VertiFL generally surpasses state-of-the-art methods in F1-score performance, while maintaining a decentralized and privacy-preserving framework.

Decentralized Federated Learning (DFL), a paradigm for managing big data in a privacy-preserved manner, is still vulnerable to poisoning attacks where malicious clients tamper with data or models. Current defense methods often assume Independently and Identically Distributed (IID) data, which is unrealistic in real-world applications. In non-IID contexts, existing defensive strategies face challenges in distinguishing between models that have been compromised and those that have been trained on heterogeneous data distributions, leading to diminished efficacy. In response, this paper proposes a framework that employs the Moving Target Defense (MTD) approach to bolster the robustness of DFL models. By continuously modifying the attack surface of the DFL system, this framework aims to mitigate poisoning attacks effectively. The proposed MTD framework includes both proactive and reactive modes, utilizing a reputation system that combines metrics of model similarity and loss, alongside various defensive techniques. Comprehensive experimental evaluations indicate that the MTD-based mechanism significantly mitigates a range of poisoning attack types across multiple datasets with different topologies.

Selective exposure, individuals' inclination to seek out information that supports their beliefs while avoiding information that contradicts them, plays an important role in the emergence of polarization. In the political domain, selective exposure is usually measured on a left-right ideology scale, ignoring finer details. Here, we combine survey and Twitter data collected during the 2022 Brazilian Presidential Election and investigate selective exposure patterns between the survey respondents and political influencers. We analyze the followship network between survey respondents and political influencers and find a multilevel community structure that reveals a hierarchical organization more complex than a simple split between left and right. Moreover, depending on the level we consider, we find different associations between network indices of exposure patterns and 189 individual attributes of the survey respondents. For example, at finer levels, the number of influencer communities a survey respondent follows is associated with several factors, such as demographics, news consumption frequency, and incivility perception. In comparison, only their political ideology is a significant factor at coarser levels. Our work demonstrates that measuring selective exposure at a single level, such as left and right, misses important information necessary to capture this phenomenon correctly.

Federated Learning (FL) has emerged as a promising approach to address privacy concerns inherent in Machine Learning (ML) practices. However, conventional FL methods, particularly those following the Centralized FL (CFL) paradigm, utilize a central server for global aggregation, which exhibits limitations such as bottleneck and single point of failure. To address these issues, the Decentralized FL (DFL) paradigm has been proposed, which removes the client-server boundary and enables all participants to engage in model training and aggregation tasks. Nevertheless, as CFL, DFL remains vulnerable to adversarial attacks, notably poisoning attacks that undermine model performance. While existing research on model robustness has predominantly focused on CFL, there is a noteworthy gap in understanding the model robustness of the DFL paradigm. In this paper, a thorough review of poisoning attacks targeting the model robustness in DFL systems, as well as their corresponding countermeasures, are presented. Additionally, a solution called DART is proposed to evaluate the robustness of DFL models, which is implemented and integrated into a DFL platform. Through extensive experiments, this paper compares the behavior of CFL and DFL under diverse poisoning attacks, pinpointing key factors affecting attack spread and effectiveness within the DFL. It also evaluates the performance of different defense mechanisms and investigates whether defense mechanisms designed for CFL are compatible with DFL. The empirical results provide insights into research challenges and suggest ways to improve the robustness of DFL models for future research.

Prompting and Multiple Choices Questions (MCQ) have become the preferred approach to assess the capabilities of Large Language Models (LLMs), due to their ease of manipulation and evaluation. Such experimental appraisals have pointed toward the LLMs' apparent ability to perform causal reasoning or to grasp uncertainty. In this paper, we investigate whether these abilities are measurable outside of tailored prompting and MCQ by reformulating these issues as direct text completion - the foundation of LLMs. To achieve this goal, we define scenarios with multiple possible outcomes and we compare the prediction made by the LLM through prompting (their Stated Answer) to the probability distributions they compute over these outcomes during next token prediction (their Revealed Belief). Our findings suggest that the Revealed Belief of LLMs significantly differs from their Stated Answer and hint at multiple biases and misrepresentations that their beliefs may yield in many scenarios and outcomes. As text completion is at the core of LLMs, these results suggest that common evaluation methods may only provide a partial picture and that more research is needed to assess the extent and nature of their capabilities.

Bluesky is a nascent Twitter-like and decentralized social media network with novel features and unprecedented data access. This paper provides a characterization of its interaction network, studying the political leaning, polarization, network structure, and algorithmic curation mechanisms of five million users. The dataset spans from the website's first release in February of 2023 to May of 2024. We investigate the replies, likes, reposts, and follows layers of the Bluesky network. We find that all networks are characterized by heavy-tailed distributions, high clustering, and short connection paths, similar to other larger social networks. Bluesky introduced feeds - algorithmic content recommenders created for and by users. We analyze all feeds and find that while a large number of custom feeds have been created, users' uptake of them appears to be limited. We analyze the hyperlinks shared by Bluesky's users and find no evidence of polarization in terms of the political leaning of the news sources they share. They share predominantly left-center news sources and little to no links associated with questionable news sources. In contrast to the homogeneous political ideology, we find significant issues-based divergence by studying opinions related to the Israel-Palestine conflict. Two clear homophilic clusters emerge: Pro-Palestinian voices outnumber pro-Israeli users, and the proportion has increased. We conclude by claiming that Bluesky-for all its novel features - is very similar in its network structure to existing and larger social media sites and provides unprecedented research opportunities for social scientists, network scientists, and political scientists alike.

In the current cybersecurity landscape, protecting military devices such as communication and battlefield management systems against sophisticated cyber attacks is crucial. Malware exploits vulnerabilities through stealth methods, often evading traditional detection mechanisms such as software signatures. The application of ML/DL in vulnerability detection has been extensively explored in the literature. However, current ML/DL vulnerability detection methods struggle with understanding the context and intent behind complex attacks. Integrating large language models (LLMs) with system call analysis offers a promising approach to enhance malware detection. This work presents a novel framework leveraging LLMs to classify malware based on system call data. The framework uses transfer learning to adapt pre-trained LLMs for malware detection. By retraining LLMs on a dataset of benign and malicious system calls, the models are refined to detect signs of malware activity. Experiments with a dataset of over 1TB of system calls demonstrate that models with larger context sizes, such as BigBird and Longformer, achieve superior accuracy and F1-Score of approximately 0.86. The results highlight the importance of context size in improving detection rates and underscore the trade-offs between computational complexity and performance. This approach shows significant potential for real-time detection in high-stakes environments, offering a robust solution to evolving cyber threats., Comment: Submitted to IEEE MILCOM 2024

Threat modeling is a popular method to securely develop systems by achieving awareness of potential areas of future damage caused by adversaries. However, threat modeling for systems relying on Artificial Intelligence is still not well explored. While conventional threat modeling methods and tools did not address AI-related threats, research on this amalgamation still lacks solutions capable of guiding and automating the process, as well as providing evidence that the methods hold up in practice. Consequently, this paper presents ThreatFinderAI, an approach and tool providing guidance and automation to model AI-related assets, threats, countermeasures, and quantify residual risks. To evaluate the practicality of the approach, participants were tasked to recreate a threat model developed by cybersecurity experts of an AI-based healthcare platform. Secondly, the approach was used to identify and discuss strategic risks in an LLM-based application through a case study. Overall, the solution's usability was well-perceived and effectively supports threat identification and risk discussion.

Although ransomware has received broad attention in media and research, this evolving threat vector still poses a systematic threat. Related literature has explored their detection using various approaches leveraging Machine and Deep Learning. While these approaches are effective in detecting malware, they do not answer how to use this intelligence to protect against threats, raising concerns about their applicability in a hostile environment. Solutions that focus on mitigation rarely explore how to prevent and not just alert or halt its execution, especially when considering Linux-based samples. This paper presents GuardFS, a file system-based approach to investigate the integration of detection and mitigation of ransomware. Using a bespoke overlay file system, data is extracted before files are accessed. Models trained on this data are used by three novel defense configurations that obfuscate, delay, or track access to the file system. The experiments on GuardFS test the configurations in a reactive setting. The results demonstrate that although data loss cannot be completely prevented, it can be significantly reduced. Usability and performance analysis demonstrate that the defense effectiveness of the configurations relates to their impact on resource consumption and usability.

The dark web has become notorious for its association with illicit activities and there is a growing need for systems to automate the monitoring of this space. This paper proposes an end-to-end scalable architecture for the early identification of new Tor sites and the daily analysis of their content. The solution is built using an Open Source Big Data stack for data serving with Kubernetes, Kafka, Kubeflow, and MinIO, continuously discovering onion addresses in different sources (threat intelligence, code repositories, web-Tor gateways, and Tor repositories), downloading the HTML from Tor and deduplicating the content using MinHash LSH, and categorizing with the BERTopic modeling (SBERT embedding, UMAP dimensionality reduction, HDBSCAN document clustering and c-TF-IDF topic keywords). In 93 days, the system identified 80,049 onion services and characterized 90% of them, addressing the challenge of Tor volatility. A disproportionate amount of repeated content is found, with only 6.1% unique sites. From the HTML files of the dark sites, 31 different low-topics are extracted, manually labeled, and grouped into 11 high-level topics. The five most popular included sexual and violent content, repositories, search engines, carding, cryptocurrencies, and marketplaces. During the experiments, we identified 14 sites with 13,946 clones that shared a suspiciously similar mirroring rate per day, suggesting an extensive common phishing network. Among the related works, this study is the most representative characterization of onion services based on topics to date.

Artificial intelligence (AI) plays a pivotal role in various sectors, influencing critical decision-making processes in our daily lives. Within the AI landscape, novel AI paradigms, such as Federated Learning (FL), focus on preserving data privacy while collaboratively training AI models. In such a context, a group of experts from the European Commission (AI-HLEG) has identified sustainable AI as one of the key elements that must be considered to provide trustworthy AI. While existing literature offers several taxonomies and solutions for assessing the trustworthiness of FL models, a significant gap exists in considering sustainability and the carbon footprint associated with FL. Thus, this work introduces the sustainability pillar to the most recent and comprehensive trustworthy FL taxonomy, making this work the first to address all AI-HLEG requirements. The sustainability pillar assesses the FL system environmental impact, incorporating notions and metrics for hardware efficiency, federation complexity, and energy grid carbon intensity. Then, this work designs and implements an algorithm for evaluating the trustworthiness of FL models by incorporating the sustainability pillar. Extensive evaluations with the FederatedScope framework and various scenarios varying federation participants, complexities, hardware, and energy grids demonstrate the usefulness of the proposed solution.

Misinformation and disinformation are growing threats in the digital age, spreading rapidly across languages and borders. This paper investigates the prevalence and dynamics of multilingual misinformation through an analysis of over 250,000 unique fact-checks spanning 95 languages. First, we find that while the majority of misinformation claims are only fact-checked once, 11.7%, corresponding to more than 21,000 claims, are checked multiple times. Using fact-checks as a proxy for the spread of misinformation, we find 33% of repeated claims cross linguistic boundaries, suggesting that some misinformation permeates language barriers. However, spreading patterns exhibit strong homophily, with misinformation more likely to spread within the same language. To study the evolution of claims over time and mutations across languages, we represent fact-checks with multilingual sentence embeddings and cluster semantically similar claims. We analyze the connected components and shortest paths connecting different versions of a claim finding that claims gradually drift over time and undergo greater alteration when traversing languages. Overall, this novel investigation of multilingual misinformation provides key insights. It quantifies redundant fact-checking efforts, establishes that some claims diffuse across languages, measures linguistic homophily, and models the temporal and cross-lingual evolution of claims. The findings advocate for expanded information sharing between fact-checkers globally while underscoring the importance of localized verification.

Automated fact-checking, using machine learning to verify claims, has grown vital as misinformation spreads beyond human fact-checking capacity. Large Language Models (LLMs) like GPT-4 are increasingly trusted to write academic papers, lawsuits, and news articles and to verify information, emphasizing their role in discerning truth from falsehood and the importance of being able to verify their outputs. Understanding the capacities and limitations of LLMs in fact-checking tasks is therefore essential for ensuring the health of our information ecosystem. Here, we evaluate the use of LLM agents in fact-checking by having them phrase queries, retrieve contextual data, and make decisions. Importantly, in our framework, agents explain their reasoning and cite the relevant sources from the retrieved context. Our results show the enhanced prowess of LLMs when equipped with contextual information. GPT-4 outperforms GPT-3, but accuracy varies based on query language and claim veracity. While LLMs show promise in fact-checking, caution is essential due to inconsistent accuracy. Our investigation calls for further research, fostering a deeper comprehension of when agents succeed and when they fail.

The growing concern over malicious attacks targeting the robustness of both Centralized and Decentralized Federated Learning (FL) necessitates novel defensive strategies. In contrast to the centralized approach, Decentralized FL (DFL) has the advantage of utilizing network topology and local dataset information, enabling the exploration of Moving Target Defense (MTD) based approaches. This work presents a theoretical analysis of the influence of network topology on the robustness of DFL models. Drawing inspiration from these findings, a three-stage MTD-based aggregation protocol, called Voyager, is proposed to improve the robustness of DFL models against poisoning attacks by manipulating network topology connectivity. Voyager has three main components: an anomaly detector, a network topology explorer, and a connection deployer. When an abnormal model is detected in the network, the topology explorer responds strategically by forming connections with more trustworthy participants to secure the model. Experimental evaluations show that Voyager effectively mitigates various poisoning attacks without imposing significant resource and computational burdens on participants. These findings highlight the proposed reactive MTD as a potent defense mechanism in the context of DFL.

Decentralized Federated Learning (DFL) emerges as an innovative paradigm to train collaborative models, addressing the single point of failure limitation. However, the security and trustworthiness of FL and DFL are compromised by poisoning attacks, negatively impacting its performance. Existing defense mechanisms have been designed for centralized FL and they do not adequately exploit the particularities of DFL. Thus, this work introduces Sentinel, a defense strategy to counteract poisoning attacks in DFL. Sentinel leverages the accessibility of local data and defines a three-step aggregation protocol consisting of similarity filtering, bootstrap validation, and normalization to safeguard against malicious model updates. Sentinel has been evaluated with diverse datasets and data distributions. Besides, various poisoning attack types and threat levels have been verified. The results improve the state-of-the-art performance against both untargeted and targeted poisoning attacks when data follows an IID (Independent and Identically Distributed) configuration. Besides, under non-IID configuration, it is analyzed how performance degrades both for Sentinel and other state-of-the-art robust aggregation methods.

Recent research has shown that the integration of Reinforcement Learning (RL) with Moving Target Defense (MTD) can enhance cybersecurity in Internet-of-Things (IoT) devices. Nevertheless, the practicality of existing work is hindered by data privacy concerns associated with centralized data processing in RL, and the unsatisfactory time needed to learn right MTD techniques that are effective against a rising number of heterogeneous zero-day attacks. Thus, this work presents CyberForce, a framework that combines Federated and Reinforcement Learning (FRL) to collaboratively and privately learn suitable MTD techniques for mitigating zero-day attacks. CyberForce integrates device fingerprinting and anomaly detection to reward or penalize MTD mechanisms chosen by an FRL-based agent. The framework has been deployed and evaluated in a scenario consisting of ten physical devices of a real IoT platform affected by heterogeneous malware samples. A pool of experiments has demonstrated that CyberForce learns the MTD technique mitigating each attack faster than existing RL-based centralized approaches. In addition, when various devices are exposed to different attacks, CyberForce benefits from knowledge transfer, leading to enhanced performance and reduced learning time in comparison to recent works. Finally, different aggregation algorithms used during the agent learning process provide CyberForce with notable robustness to malicious attacks.

The ongoing debate surrounding the impact of the Internet Research Agency s (IRA) social media campaign during the 2016 U.S. presidential election has largely overshadowed the involvement of other actors. Our analysis brings to light a substantial group of suspended Twitter users, outnumbering the IRA user group by a factor of 60, who align with the ideologies of the IRA campaign. Our study demonstrates that this group of suspended Twitter accounts significantly influenced individuals categorized as undecided or weak supporters, potentially with the aim of swaying their opinions, as indicated by Granger causality., Comment: 7 Tables, 5 Figures

The rise of Decentralized Federated Learning (DFL) has enabled the training of machine learning models across federated participants, fostering decentralized model aggregation and reducing dependence on a server. However, this approach introduces unique communication security challenges that have yet to be thoroughly addressed in the literature. These challenges primarily originate from the decentralized nature of the aggregation process, the varied roles and responsibilities of the participants, and the absence of a central authority to oversee and mitigate threats. Addressing these challenges, this paper first delineates a comprehensive threat model focused on DFL communications. In response to these identified risks, this work introduces a security module to counter communication-based attacks for DFL platforms. The module combines security techniques such as symmetric and asymmetric encryption with Moving Target Defense (MTD) techniques, including random neighbor selection and IP/port switching. The security module is implemented in a DFL platform, Fedstellar, allowing the deployment and monitoring of the federation. A DFL scenario with physical and virtual deployments have been executed, encompassing three security configurations: (i) a baseline without security, (ii) an encrypted configuration, and (iii) a configuration integrating both encryption and MTD techniques. The effectiveness of the security module is validated through experiments with the MNIST dataset and eclipse attacks. The results showed an average F1 score of 95%, with the most secure configuration resulting in CPU usage peaking at 68% (+-9%) in virtual deployments and network traffic reaching 480.8 MB (+-18 MB), effectively mitigating risks associated with eavesdropping or eclipse attacks.

Ransomware has remained one of the most notorious threats in the cybersecurity field. Moving Target Defense (MTD) has been proposed as a novel paradigm for proactive defense. Although various approaches leverage MTD, few of them rely on the operating system and, specifically, the file system, thereby making them dependent on other computing devices. Furthermore, existing ransomware defense techniques merely replicate or detect attacks, without preventing them. Thus, this paper introduces the MTFS overlay file system and the design and implementation of three novel MTD techniques implemented on top of it. One delaying attackers, one trapping recursive directory traversal, and another one hiding file types. The effectiveness of the techniques are shown in two experiments. First, it is shown that the techniques can delay and mitigate ransomware on real IoT devices. Secondly, in a broader scope, the solution was confronted with 14 ransomware samples, highlighting that it can save 97% of the files.

Cybersecurity solutions have shown promising performance when detecting ransomware samples that use fixed algorithms and encryption rates. However, due to the current explosion of Artificial Intelligence (AI), sooner than later, ransomware (and malware in general) will incorporate AI techniques to intelligently and dynamically adapt its encryption behavior to be undetected. It might result in ineffective and obsolete cybersecurity solutions, but the literature lacks AI-powered ransomware to verify it. Thus, this work proposes RansomAI, a Reinforcement Learning-based framework that can be integrated into existing ransomware samples to adapt their encryption behavior and stay stealthy while encrypting files. RansomAI presents an agent that learns the best encryption algorithm, rate, and duration that minimizes its detection (using a reward mechanism and a fingerprinting intelligent detection system) while maximizing its damage function. The proposed framework was validated in a ransomware, Ransomware-PoC, that infected a Raspberry Pi 4, acting as a crowdsensor. A pool of experiments with Deep Q-Learning and Isolation Forest (deployed on the agent and detection system, respectively) has demonstrated that RansomAI evades the detection of Ransomware-PoC affecting the Raspberry Pi 4 in a few minutes with >90% accuracy.

In 2016, Google proposed Federated Learning (FL) as a novel paradigm to train Machine Learning (ML) models across the participants of a federation while preserving data privacy. Since its birth, Centralized FL (CFL) has been the most used approach, where a central entity aggregates participants' models to create a global one. However, CFL presents limitations such as communication bottlenecks, single point of failure, and reliance on a central server. Decentralized Federated Learning (DFL) addresses these issues by enabling decentralized model aggregation and minimizing dependency on a central entity. Despite these advances, current platforms training DFL models struggle with key issues such as managing heterogeneous federation network topologies. To overcome these challenges, this paper presents Fedstellar, a platform extended from p2pfl library and designed to train FL models in a decentralized, semi-decentralized, and centralized fashion across diverse federations of physical or virtualized devices. The Fedstellar implementation encompasses a web application with an interactive graphical interface, a controller for deploying federations of nodes using physical or virtual devices, and a core deployed on each device which provides the logic needed to train, aggregate, and communicate in the network. The effectiveness of the platform has been demonstrated in two scenarios: a physical deployment involving single-board devices such as Raspberry Pis for detecting cyberattacks, and a virtualized deployment comparing various FL approaches in a controlled environment using MNIST and CIFAR-10 datasets. In both scenarios, Fedstellar demonstrated consistent performance and adaptability, achieving F1 scores of 91%, 98%, and 91.2% using DFL for detecting cyberattacks and classifying MNIST and CIFAR-10, respectively, reducing training time by 32% compared to centralized approaches.

The proliferation of the Internet of Things (IoT) has led to the emergence of crowdsensing applications, where a multitude of interconnected devices collaboratively collect and analyze data. Ensuring the authenticity and integrity of the data collected by these devices is crucial for reliable decision-making and maintaining trust in the system. Traditional authentication methods are often vulnerable to attacks or can be easily duplicated, posing challenges to securing crowdsensing applications. Besides, current solutions leveraging device behavior are mostly focused on device identification, which is a simpler task than authentication. To address these issues, an individual IoT device authentication framework based on hardware behavior fingerprinting and Transformer autoencoders is proposed in this work. This solution leverages the inherent imperfections and variations in IoT device hardware to differentiate between devices with identical specifications. By monitoring and analyzing the behavior of key hardware components, such as the CPU, GPU, RAM, and Storage on devices, unique fingerprints for each device are created. The performance samples are considered as time series data and used to train outlier detection transformer models, one per device and aiming to model its normal data distribution. Then, the framework is validated within a spectrum crowdsensing system leveraging Raspberry Pi devices. After a pool of experiments, the model from each device is able to individually authenticate it between the 45 devices employed for validation. An average True Positive Rate (TPR) of 0.74+-0.13 and an average maximum False Positive Rate (FPR) of 0.06+-0.09 demonstrate the effectiveness of this approach in enhancing authentication, security, and trust in crowdsensing applications.

The rapid expansion of the Internet of Things (IoT) and Edge Computing has presented challenges for centralized Machine and Deep Learning (ML/DL) methods due to the presence of distributed data silos that hold sensitive information. To address concerns regarding data privacy, collaborative and privacy-preserving ML/DL techniques like Federated Learning (FL) have emerged. However, ensuring data privacy and performance alone is insufficient since there is a growing need to establish trust in model predictions. Existing literature has proposed various approaches on trustworthy ML/DL (excluding data privacy), identifying robustness, fairness, explainability, and accountability as important pillars. Nevertheless, further research is required to identify trustworthiness pillars and evaluation metrics specifically relevant to FL models, as well as to develop solutions that can compute the trustworthiness level of FL models. This work examines the existing requirements for evaluating trustworthiness in FL and introduces a comprehensive taxonomy consisting of six pillars (privacy, robustness, fairness, explainability, accountability, and federation), along with over 30 metrics for computing the trustworthiness of FL models. Subsequently, an algorithm named FederatedTrust is designed based on the pillars and metrics identified in the taxonomy to compute the trustworthiness score of FL models. A prototype of FederatedTrust is implemented and integrated into the learning process of FederatedScope, a well-established FL framework. Finally, five experiments are conducted using different configurations of FederatedScope to demonstrate the utility of FederatedTrust in computing the trustworthiness of FL models. Three experiments employ the FEMNIST dataset, and two utilize the N-BaIoT dataset considering a real-world IoT security use case.

In the last years, the number of IoT devices deployed has suffered an undoubted explosion, reaching the scale of billions. However, some new cybersecurity issues have appeared together with this development. Some of these issues are the deployment of unauthorized devices, malicious code modification, malware deployment, or vulnerability exploitation. This fact has motivated the requirement for new device identification mechanisms based on behavior monitoring. Besides, these solutions have recently leveraged Machine and Deep Learning techniques due to the advances in this field and the increase in processing capabilities. In contrast, attackers do not stay stalled and have developed adversarial attacks focused on context modification and ML/DL evaluation evasion applied to IoT device identification solutions. This work explores the performance of hardware behavior-based individual device identification, how it is affected by possible context- and ML/DL-focused attacks, and how its resilience can be improved using defense techniques. In this sense, it proposes an LSTM-CNN architecture based on hardware performance behavior for individual device identification. Then, previous techniques have been compared with the proposed architecture using a hardware performance dataset collected from 45 Raspberry Pi devices running identical software. The LSTM-CNN improves previous solutions achieving a +0.96 average F1-Score and 0.8 minimum TPR for all devices. Afterward, context- and ML/DL-focused adversarial attacks were applied against the previous model to test its robustness. A temperature-based context attack was not able to disrupt the identification. However, some ML/DL state-of-the-art evasion attacks were successful. Finally, adversarial training and model distillation defense techniques are selected to improve the model resilience to evasion attacks, without degrading its performance.

Cybercriminals are moving towards zero-day attacks affecting resource-constrained devices such as single-board computers (SBC). Assuming that perfect security is unrealistic, Moving Target Defense (MTD) is a promising approach to mitigate attacks by dynamically altering target attack surfaces. Still, selecting suitable MTD techniques for zero-day attacks is an open challenge. Reinforcement Learning (RL) could be an effective approach to optimize the MTD selection through trial and error, but the literature fails when i) evaluating the performance of RL and MTD solutions in real-world scenarios, ii) studying whether behavioral fingerprinting is suitable for representing SBC's states, and iii) calculating the consumption of resources in SBC. To improve these limitations, the work at hand proposes an online RL-based framework to learn the correct MTD mechanisms mitigating heterogeneous zero-day attacks in SBC. The framework considers behavioral fingerprinting to represent SBCs' states and RL to learn MTD techniques that mitigate each malicious state. It has been deployed on a real IoT crowdsensing scenario with a Raspberry Pi acting as a spectrum sensor. More in detail, the Raspberry Pi has been infected with different samples of command and control malware, rootkits, and ransomware to later select between four existing MTD techniques. A set of experiments demonstrated the suitability of the framework to learn proper MTD techniques mitigating all attacks (except a harmfulness rootkit) while consuming <1 MB of storage and utilizing <55% CPU and <80% RAM.

In recent years, Federated Learning (FL) has gained relevance in training collaborative models without sharing sensitive data. Since its birth, Centralized FL (CFL) has been the most common approach in the literature, where a central entity creates a global model. However, a centralized approach leads to increased latency due to bottlenecks, heightened vulnerability to system failures, and trustworthiness concerns affecting the entity responsible for the global model creation. Decentralized Federated Learning (DFL) emerged to address these concerns by promoting decentralized model aggregation and minimizing reliance on centralized architectures. However, despite the work done in DFL, the literature has not (i) studied the main aspects differentiating DFL and CFL; (ii) analyzed DFL frameworks to create and evaluate new solutions; and (iii) reviewed application scenarios using DFL. Thus, this article identifies and analyzes the main fundamentals of DFL in terms of federation architectures, topologies, communication mechanisms, security approaches, and key performance indicators. Additionally, the paper at hand explores existing mechanisms to optimize critical DFL fundamentals. Then, the most relevant features of the current DFL frameworks are reviewed and compared. After that, it analyzes the most used DFL application scenarios, identifying solutions based on the fundamentals and frameworks previously defined. Finally, the evolution of existing DFL solutions is studied to provide a list of trends, lessons learned, and open challenges.

Deep Neural Networks are being extensively used in communication systems and Automatic Modulation Classification (AMC) in particular. However, they are very susceptible to small adversarial perturbations that are carefully crafted to change the network decision. In this work, we build on knowledge distillation ideas and adversarial training in order to build more robust AMC systems. We first outline the importance of the quality of the training data in terms of accuracy and robustness of the model. We then propose to use the Maximum Likelihood function, which could solve the AMC problem in offline settings, to generate better training labels. Those labels teach the model to be uncertain in challenging conditions, which permits to increase the accuracy, as well as the robustness of the model when combined with adversarial training. Interestingly, we observe that this increase in performance transfers to online settings, where the Maximum Likelihood function cannot be used in practice. Overall, this work highlights the potential of learning to be uncertain in difficult scenarios, compared to directly removing label noise.

Federated learning (FL) allows participants to collaboratively train machine and deep learning models while protecting data privacy. However, the FL paradigm still presents drawbacks affecting its trustworthiness since malicious participants could launch adversarial attacks against the training process. Related work has studied the robustness of horizontal FL scenarios under different attacks. However, there is a lack of work evaluating the robustness of decentralized vertical FL and comparing it with horizontal FL architectures affected by adversarial attacks. Thus, this work proposes three decentralized FL architectures, one for horizontal and two for vertical scenarios, namely HoriChain, VertiChain, and VertiComb. These architectures present different neural networks and training protocols suitable for horizontal and vertical scenarios. Then, a decentralized, privacy-preserving, and federated use case with non-IID data to classify handwritten digits is deployed to evaluate the performance of the three architectures. Finally, a set of experiments computes and compares the robustness of the proposed architectures when they are affected by different data poisoning based on image watermarks and gradient poisoning adversarial attacks. The experiments show that even though particular configurations of both attacks can destroy the classification performance of the architectures, HoriChain is the most robust one.

Malware affecting Internet of Things (IoT) devices is rapidly growing due to the relevance of this paradigm in real-world scenarios. Specialized literature has also detected a trend towards multi-purpose malware able to execute different malicious actions such as remote control, data leakage, encryption, or code hiding, among others. Protecting IoT devices against this kind of malware is challenging due to their well-known vulnerabilities and limitation in terms of CPU, memory, and storage. To improve it, the moving target defense (MTD) paradigm was proposed a decade ago and has shown promising results, but there is a lack of IoT MTD solutions dealing with multi-purpose malware. Thus, this work proposes four MTD mechanisms changing IoT devices' network, data, and runtime environment to mitigate multi-purpose malware. Furthermore, it presents a lightweight and IoT-oriented MTD framework to decide what, when, and how the MTD mechanisms are deployed. Finally, the efficiency and effectiveness of the framework and MTD mechanisms are evaluated in a real-world scenario with one IoT spectrum sensor affected by multi-purpose malware.

In today's computing environment, where Artificial Intelligence (AI) and data processing are moving toward the Internet of Things (IoT) and Edge computing paradigms, benchmarking resource-constrained devices is a critical task to evaluate their suitability and performance. Between the employed devices, Single-Board Computers arise as multi-purpose and affordable systems. The literature has explored Single-Board Computers performance when running high-level benchmarks specialized in particular application scenarios, such as AI or medical applications. However, lower-level benchmarking applications and datasets are needed to enable new Edge-based AI solutions for network, system and service management based on device and component performance, such as individual device identification. Thus, this paper presents LwHBench, a low-level hardware benchmarking application for Single-Board Computers that measures the performance of CPU, GPU, Memory and Storage taking into account the component constraints in these types of devices. LwHBench has been implemented for Raspberry Pi devices and run for 100 days on a set of 45 devices to generate an extensive dataset that allows the usage of AI techniques in scenarios where performance data can help in the device management process. Besides, to demonstrate the inter-scenario capability of the dataset, a series of AI-enabled use cases about device identification and context impact on performance are presented as exploration of the published data. Finally, the benchmark application has been adapted and applied to an agriculture-focused scenario where three RockPro64 devices are present.