IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Contributions of digital twins services to the implementation of the circular economy
No full textPart 4: Fostering Collaborative and Interoperable Digital Models for Digital Twins: CasesInternational audienceThis article examines the role of digital twins in advancing regeneration processes within the circular economy. We focus on the capabilities of digital twins to model product life cycles and improve sustainability by supporting regeneration strategies. Our results show that digital twins facilitate accurate simulation and prediction of product lifecycle phases, which is essential for decision-making. Notably, real-world implementation of digital twins has improved resource management and operational efficiency, proving that they can transform sustainability practices in manufacturing industries. This study highlights not only the operational benefits of digital twins, but also their strategic role in promoting environmental sustainability through improved resource efficiency. In addition, it identifies opportunities and challenges, which provides a model for using digital twins to obtain environmental and economic results in the manufacturing industry
FCGAN: Spectral Convolutions via FFT for Channel-Wide Receptive Field in Generative Adversarial Networks
No full textPart 1: Deep LearningInternational audienceIn this paper, we propose the Fast Fourier Convolution Generative Adversarial Network (FCGAN). This novel approach employs convolutions in the frequency domain to enable the network to operate with a channel-wide receptive field. Due to small receptive fields, traditional convolution-based GANs struggle to capture structural and geometric patterns. Our method applies Fast Fourier Convolutions (FFCs), which use Fourier Transforms to operate in the spectral domain, affecting the feature input globally. We show that this new hallmark doesn’t hinder the stability of adversarial training even for channel-wide convolutions. Our experiments further support the claim that Fourier features are lightweight replacements for self-attention, allowing the network to learn global information from early layers. We present qualitative and quantitative results to demonstrate that the proposed FCGAN achieves results comparable to state-of-the-art approaches of similar depth and parameter count. Moreover, in larger image dimensions, using FFCs instead of self-attention allows for batch sizes up to twice as large and iterations up to 26% faster
Risk Assessment of COVID-19 Transmission on Cruise Ships Using Fuzzy Rules
No full textPart 3: Data Mining/ModelingInternational audienceCruise ships constitute a popular means of vacationing for millions of people each year. However, due to the on-board conditions, e.g., densely populated areas, highly transmissible respiratory diseases, such as COVID-19, are a common cause of outbreaks. Hence, accurate assessment of the transmission risk (TR) is crucial. Recent approaches focus on long-term forecasting of such events; however, the limited availability and inconsistency of relevant data poses a challenge for developing short-term and data-driven methods. To this end, this work proposes a novel short-term knowledge-based method implemented through fuzzy rules for assessing the TR in cruise ships. The use of fuzzy rules, developed by domain experts and information extracted from the literature, assists in dealing with the data limitations. In contrast to previous approaches, the proposed method considers information deriving from various sensors and the ship information system in accord with a recently proposed smart ship design. Moreover, the fuzzy TR assessment estimates the confidence of an inferred decision, quantifying the uncertainty regarding its results. Evaluation via agent-based simulations demonstrates the effectiveness of the proposed method across different scenarios
C-XGBoost: A Tree Boosting Model for Causal Effect Estimation
No full textPart 1: Deep LearningInternational audienceCausal effect estimation aims at estimating the Average Treatment Effect as well as the Conditional Average Treatment Effect of a treatment to an outcome from the available data. This knowledge is important in many safety-critical domains, where it often needs to be extracted from observational data. In this work, we propose a new causal inference model, named C-XGBoost, for the prediction of potential outcomes. The motivation of our approach is to exploit the superiority of tree-based models for handling tabular data together with the notable property of causal inference neural network-based models to learn representations that are useful for estimating the outcome for both the treatment and non-treatment cases. The proposed model also inherits the considerable advantages of XGBoost model such as efficiently handling features with missing values requiring minimum preprocessing effort, as well as it is equipped with regularization techniques to avoid overfitting/bias. Furthermore, we propose a new loss function for efficiently training the proposed causal inference model. The experimental analysis, which is based on the performance profiles of Dolan and Moré as well as on post-hoc and non-parametric statistical tests, provide strong evidence about the effectiveness of the proposed approach
Multi-adaptive Neural Modelling of the Interplay of Changing Organisational Contexts, Epigenetics, and Personality Traits in the Development of Burnout
No full textPart 2: Graphs/Neural Networks/Machine LearningInternational audienceThis research addresses the influence of a changing organisational context and the big five personality traits on the three main characterising elements of a burnout. A computational analysis is contributed based on an adaptive network modeling approach. The simulation results show how someone who is high in personality traits such as agreeableness, openness, extraversion, conscientiousness, and highly sensitive for neuroticism, is vulnerable to reach a burnout level in all dimensions whenever the organisational context is changing in a less favorable direction. It is also shown how therapy alone may not be sufficient as a long-term treatment
ICT Systems Security and Privacy Protection: 39th IFIP International Conference, SEC 2024, Edinburgh, UK, June 12–14, 2024 Proceedings
No full textInternational audienceBook Front Matter of AICT 71
Rasd: Semantic Shift Detection and Adaptation for Network Intrusion Detection
No full textInternational audienceNetwork Intrusion Detection Systems (NIDSs) based on Deep Neural Network have demonstrated impressive performance in multi-class, closed-world settings, where training and test data follow the same distribution. However, when deployed in real networks, these systems have a limited ability to detect novel attacks which do not belong to already known classes. In this work, we aim to tackle semantic shift, that is the emergence of unknown classes, by proposing a two-phase approach to detect new classes and integrate them into the classification model, while minimising the need for human intervention. While contrastive learning is a promising techniques to tackle semantic shift, it has high computational cost and it is sensitive to imbalanced data. We propose a novel contrastive learning approach based on synthetic centroids which has low computational cost and is robust to class imbalance, making it suitable for application to NIDS. To integrate the shifted samples in the existing model, we also design a novel adaptation method that combines manual labeling and pseudo-labeling to reduce labeling costs. We evaluate our system, Rasd, on two NIDS datasets, finding it excels in both detection and adaptation. For example Rasd improves on the nearest detection baseline F1-score by 6.83% for IDS 2017 and 19.21% for IDS 2018
Multi-method Approach to Human Expertise, Automation, and Artificial Intelligence for Vulnerability Management
No full textInternational audienceFast-growing digital trends have driven growth in the threat landscape of cyber-attacks, pushing unprecedented burdens on organizations to manage vulnerabilities effectively. This study investigated two years of complex relationships between human expertise and technological solutions in the domain of cybersecurity vulnerability management (VM) for a leading fast-moving consumer goods (FMCG) company operating internationally in multiple countries, leveraging both on-premises and cloud infrastructure. This study introduces the tensions arising from this duality. Rooted in sociotechnical systems theory (STS), actor-network theory (ANT), and resource-based view (RBV), this research bridges the gap between technological reliance and human interpretative skills, which are two dominant but often disconnected aspects of VM. This paper highlights the benefit of VM that results from a symbiotic relationship between humans and technology, emphasizing how artificial intelligence (AI) and automation can mitigate the limitations of human-centric approaches and how humans can address the technological contextual limitations, resulting in a win-win approach. The findings set the orientation for a nascent stream of academic research on the relationship between humans and AI in vulnerability management
Towards an Unbiased Classification of Chest X-ray Images Using a RL Powered ACGAN Framework
No full textPart 1: Biomedical/ClassificationInternational audienceComputer-aided diagnosis systems are invaluable tools for healthcare providers given the overwhelming volume of medical data at their disposal. However, a significant challenge of these systems is the existence of bias in their diagnostic outcomes, particularly affecting certain protected groups who are more susceptible to receiving incorrect diagnoses. In this paper, we investigate bias mitigation strategies, leveraging the discriminator of auxiliary conditional generative adversarial networks as well as reinforcement learning agents for the classification of chest X-ray images. Our research targets bias reduction by deploying reward functions designed to enhance the true positivity rate of the discriminator. We explore the impact of a hierarchical label distribution-based reward, a novel approach that aims to further improve bias in the diagnostic process. Through extensive evaluation and comparison, we study the disparities in true positivity rates across various different approaches. We also highlight the efficiency of each strategy in achieving more equitable diagnostic outcomes
Evaluation of Language Models for Multilabel Classification of Biomedical Texts
No full textPart 1: Biomedical/ClassificationInternational audienceThe continuous increase of data availability and the need for their utilization make it imperative to organize them into categories. Recent classification problems often involve the prediction of multiple labels simultaneously applying to a single instance. In this paper, we propose a structured approach for the implementation and evaluation of multilabel classification tasks in the context of biomedical texts. This involves selecting appropriate datasets and models, designing experiments, and defining metrics that accurately measure the models’ performance across various aspects of the task. Our results yield notable scores and conclusions for the behavior of some state-of-the-art language models in specific data. It is shown that the complexity of biomedical data and the intricacy of multilabel classification require careful consideration of these models’ capabilities to handle large label spaces, label correlations, and the nuances of biomedical language