arXiv subjects

Christos Anagnostopoulos

Publications and source records attributed to Christos Anagnostopoulos.

At least 37 records · Page 2Linked to original sources

arXiv2025

FedPhD: Federated Pruning with Hierarchical Learning of Diffusion Models

Federated Learning (FL), as a distributed learning paradigm, trains models over distributed clients' data. FL is particularly beneficial for distributed training of Diffusion Models (DMs), which are high-quality image generators that require diverse data. However, challenges such as high communication costs and data heterogeneity persist in training DMs similar to training Transformers and Convolutional Neural Networks. Limited research has addressed these issues in FL environments. To address this gap and challenges, we introduce a novel approach, FedPhD, designed to efficiently train DMs in FL environments. FedPhD leverages Hierarchical FL with homogeneity-aware model aggregation and selection policy to tackle data heterogeneity while reducing communication costs. The distributed structured pruning of FedPhD enhances computational efficiency and reduces model storage requirements in clients. Our experiments across multiple datasets demonstrate that FedPhD achieves high model performance regarding Fréchet Inception Distance (FID) scores while reducing communication costs by up to $88\%$. FedPhD outperforms baseline methods achieving at least a $34\%$ improvement in FID, while utilizing only $56\%$ of the total computation and communication resources.

cs.LG↗

arXiv2025

Distributed Temporal Graph Learning with Provenance for APT Detection in Supply Chains

Cyber supply chain, encompassing digital asserts, software, hardware, has become an essential component of modern Information and Communications Technology (ICT) provisioning. However, the growing inter-dependencies have introduced numerous attack vectors, making supply chains a prime target for exploitation. In particular, advanced persistent threats (APTs) frequently leverage supply chain vulnerabilities (SCVs) as entry points, benefiting from their inherent stealth. Current defense strategies primarly focus on prevention through blockchain for integrity assurance or detection using plain-text source code analysis in open-source software (OSS). However, these approaches overlook scenarios where source code is unavailable and fail to address detection and defense during runtime. To bridge this gap, we propose a novel approach that integrates multi-source data, constructs a comprehensive dynamic provenance graph, and detects APT behavior in real time using temporal graph learning. Given the lack of tailored datasets in both industry and academia, we also aim to simulate a custom dataset by replaying real-world supply chain exploits with multi-source monitoring.

cs.CR↗

arXiv2025

Distributed Log-driven Anomaly Detection System based on Evolving Decision Making

Effective anomaly detection from logs is crucial for enhancing cybersecurity defenses by enabling the early identification of threats. Despite advances in anomaly detection, existing systems often fall short in areas such as post-detection validation, scalability, and effective maintenance. These limitations not only hinder the detection of new threats but also impair overall system performance. To address these challenges, we propose CEDLog, a novel practical framework that integrates Elastic Weight Consolidation (EWC) for continual learning and implements distributed computing for scalable processing by integrating Apache Airflow and Dask. In CEDLog, anomalies are detected through the synthesis of Multi-layer Perceptron (MLP) and Graph Convolutional Networks (GCNs) using critical features present in event logs. Through comparisons with update strategies on large-scale datasets, we demonstrate the strengths of CEDLog, showcasing efficient updates and low false positives

cs.CR↗

arXiv2025

Artificial Intelligence-Driven Clinical Decision Support Systems

As artificial intelligence (AI) becomes increasingly embedded in healthcare delivery, this chapter explores the critical aspects of developing reliable and ethical Clinical Decision Support Systems (CDSS). Beginning with the fundamental transition from traditional statistical models to sophisticated machine learning approaches, this work examines rigorous validation strategies and performance assessment methods, including the crucial role of model calibration and decision curve analysis. The chapter emphasizes that creating trustworthy AI systems in healthcare requires more than just technical accuracy; it demands careful consideration of fairness, explainability, and privacy. The challenge of ensuring equitable healthcare delivery through AI is stressed, discussing methods to identify and mitigate bias in clinical predictive models. The chapter then delves into explainability as a cornerstone of human-centered CDSS. This focus reflects the understanding that healthcare professionals must not only trust AI recommendations but also comprehend their underlying reasoning. The discussion advances in an analysis of privacy vulnerabilities in medical AI systems, from data leakage in deep learning models to sophisticated attacks against model explanations. The text explores privacy-preservation strategies such as differential privacy and federated learning, while acknowledging the inherent trade-offs between privacy protection and model performance. This progression, from technical validation to ethical considerations, reflects the multifaceted challenges of developing AI systems that can be seamlessly and reliably integrated into daily clinical practice while maintaining the highest standards of patient care and data protection.

cs.AI↗

arXiv2025

Federated Data-Driven Kalman Filtering for State Estimation

This paper proposes a novel localization framework based on collaborative training or federated learning paradigm, for highly accurate localization of autonomous vehicles. More specifically, we build on the standard approach of KalmanNet, a recurrent neural network aiming to estimate the underlying system uncertainty of traditional Extended Kalman Filtering, and reformulate it by the adapt-then-combine concept to FedKalmanNet. The latter is trained in a distributed manner by a group of vehicles (or clients), with local training datasets consisting of vehicular location and velocity measurements, through a global server aggregation operation. The FedKalmanNet is then used by each vehicle to localize itself, by estimating the associated system uncertainty matrices (i.e, Kalman gain). Our aim is to actually demonstrate the benefits of collaborative training for state estimation in autonomous driving, over collaborative decision-making which requires rich V2X communication resources for measurement exchange and sensor fusion under real-time constraints. An extensive experimental and evaluation study conducted in CARLA autonomous driving simulator highlights the superior performance of FedKalmanNet over state-of-the-art collaborative decision-making approaches, in localizing vehicles without the need of real-time V2X communication.

cs.RO↗

arXiv2024

Unified Semantic Log Parsing and Causal Graph Construction for Attack Attribution

Multi-source logs provide a comprehensive overview of ongoing system activities, allowing for in-depth analysis to detect potential threats. A practical approach for threat detection involves explicit extraction of entity triples (subject, action, object) towards building provenance graphs to facilitate the analysis of system behavior. However, current log parsing methods mainly focus on retrieving parameters and events from raw logs while approaches based on entity extraction are limited to processing a single type of log. To address these gaps, we contribute with a novel unified framework, coined UTLParser. UTLParser adopts semantic analysis to construct causal graphs by merging multiple sub-graphs from individual log sources in labeled log dataset. It leverages domain knowledge in threat hunting such as Points of Interest. We further explore log generation delays and provide interfaces for optimized temporal graph querying. Our experiments showcase that UTLParser overcomes drawbacks of other log parsing methods. Furthermore, UTLParser precisely extracts explicit causal threat information while being compatible with enormous downstream tasks.

cs.SE↗

arXiv2024

Personalized Federated Learning for Cross-view Geo-localization

In this paper we propose a methodology combining Federated Learning (FL) with Cross-view Image Geo-localization (CVGL) techniques. We address the challenges of data privacy and heterogeneity in autonomous vehicle environments by proposing a personalized Federated Learning scenario that allows selective sharing of model parameters. Our method implements a coarse-to-fine approach, where clients share only the coarse feature extractors while keeping fine-grained features specific to local environments. We evaluate our approach against traditional centralized and single-client training schemes using the KITTI dataset combined with satellite imagery. Results demonstrate that our federated CVGL method achieves performance close to centralized training while maintaining data privacy. The proposed partial model sharing strategy shows comparable or slightly better performance than classical FL, offering significant reduced communication overhead without sacrificing accuracy. Our work contributes to more robust and privacy-preserving localization systems for autonomous vehicles operating in diverse environments

cs.CV↗

arXiv2024

Decentralized Personalized Federated Learning based on a Conditional Sparse-to-Sparser Scheme

Decentralized Federated Learning (DFL) has become popular due to its robustness and avoidance of centralized coordination. In this paradigm, clients actively engage in training by exchanging models with their networked neighbors. However, DFL introduces increased costs in terms of training and communication. Existing methods focus on minimizing communication often overlooking training efficiency and data heterogeneity. To address this gap, we propose a novel \textit{sparse-to-sparser} training scheme: DA-DPFL. DA-DPFL initializes with a subset of model parameters, which progressively reduces during training via \textit{dynamic aggregation} and leads to substantial energy savings while retaining adequate information during critical learning periods. Our experiments showcase that DA-DPFL substantially outperforms DFL baselines in test accuracy, while achieving up to $5$ times reduction in energy costs. We provide a theoretical analysis of DA-DPFL's convergence by solidifying its applicability in decentralized and personalized learning. The code is available at:https://github.com/EricLoong/da-dpfl

cs.LG↗

arXiv2024

Improving Sequential Query Recommendation with Immediate User Feedback

We propose an algorithm for next query recommendation in interactive data exploration settings, like knowledge discovery for information gathering. The state-of-the-art query recommendation algorithms are based on sequence-to-sequence learning approaches that exploit historical interaction data. Due to the supervision involved in the learning process, such approaches fail to adapt to immediate user feedback. We propose to augment the transformer-based causal language models for query recommendations to adapt to the immediate user feedback using multi-armed bandit (MAB) framework. We conduct a large-scale experimental study using log files from a popular online literature discovery service and demonstrate that our algorithm improves the per-round regret substantially, with respect to the state-of-the-art transformer-based query recommendation models, which do not make use of immediate user feedback. Our data model and source code are available at https://github.com/shampp/exp3_ss

cs.IR↗

arXiv2023

FedDIP: Federated Learning with Extreme Dynamic Pruning and Incremental Regularization

Federated Learning (FL) has been successfully adopted for distributed training and inference of large-scale Deep Neural Networks (DNNs). However, DNNs are characterized by an extremely large number of parameters, thus, yielding significant challenges in exchanging these parameters among distributed nodes and managing the memory. Although recent DNN compression methods (e.g., sparsification, pruning) tackle such challenges, they do not holistically consider an adaptively controlled reduction of parameter exchange while maintaining high accuracy levels. We, therefore, contribute with a novel FL framework (coined FedDIP), which combines (i) dynamic model pruning with error feedback to eliminate redundant information exchange, which contributes to significant performance improvement, with (ii) incremental regularization that can achieve \textit{extreme} sparsity of models. We provide convergence analysis of FedDIP and report on a comprehensive performance and comparative assessment against state-of-the-art methods using benchmark data sets and DNN models. Our results showcase that FedDIP not only controls the model sparsity but efficiently achieves similar or better performance compared to other model pruning methods adopting incremental regularization during distributed model training. The code is available at: https://github.com/EricLoong/feddip.

cs.LG↗

arXiv2023

A Design Approach and Prototype Implementation for Factory Monitoring Based on Virtual and Augmented Reality at the Edge of Industry 4.0

Virtual and augmented reality are currently enjoying a great deal of attention from the research community and the industry towards their adoption within industrial spaces and processes. However, the current design and implementation landscape is still very fluid, while the community as a whole has not yet consolidated into concrete design directions, other than basic patterns. Other open issues include the choice over a cloud or edge-based architecture when designing such systems. Within this work, we present our approach for a monitoring intervention inside a factory space utilizing both VR and AR, based primarily on edge computing, while also utilizing the cloud. We discuss its main design directions, as well as a basic ontology to aid in simple description of factory assets. In order to highlight the design aspects of our approach, we present a prototype implementation, based on a use case scenario in a factory site, within the context of the ENERMAN H2020 project.

cs.HC↗

arXiv2023

Real time enhancement of operator's ergonomics in physical human-robot collaboration scenarios using a multi-stereo camera system

In collaborative tasks where humans work alongside machines, the robot's movements and behaviour can have a significant impact on the operator's safety, health, and comfort. To address this issue, we present a multi-stereo camera system that continuously monitors the operator's posture while they work with the robot. This system uses a novel distributed fusion approach to assess the operator's posture in real-time and to help avoid uncomfortable or unsafe positions. The system adjusts the robot's movements and informs the operator of any incorrect or potentially harmful postures, reducing the risk of accidents, strain, and musculoskeletal disorders. The analysis is personalized, taking into account the unique anthropometric characteristics of each operator, to ensure optimal ergonomics. The results of our experiments show that the proposed approach leads to improved human body postures and offers a promising solution for enhancing the ergonomics of operators in collaborative tasks.

eess.IV↗

arXiv2022

Optimizing Vision Transformers for Medical Image Segmentation

For medical image semantic segmentation (MISS), Vision Transformers have emerged as strong alternatives to convolutional neural networks thanks to their inherent ability to capture long-range correlations. However, existing research uses off-the-shelf vision Transformer blocks based on linear projections and feature processing which lack spatial and local context to refine organ boundaries. Furthermore, Transformers do not generalize well on small medical imaging datasets and rely on large-scale pre-training due to limited inductive biases. To address these problems, we demonstrate the design of a compact and accurate Transformer network for MISS, CS-Unet, which introduces convolutions in a multi-stage design for hierarchically enhancing spatial and local modeling ability of Transformers. This is mainly achieved by our well-designed Convolutional Swin Transformer (CST) block which merges convolutions with Multi-Head Self-Attention and Feed-Forward Networks for providing inherent localized spatial context and inductive biases. Experiments demonstrate CS-Unet without pre-training outperforms other counterparts by large margins on multi-organ and cardiac datasets with fewer parameters and achieves state-of-the-art performance. Our code is available at Github.

cs.CV↗

arXiv2022

A Civil Protection Early Warning System to Improve the Resilience of Adriatic-Ionian Territories to Natural and Man-made Risk

We are currently witnessing an increased occurrence of extreme weather events, causing a great deal of disruption and distress across the globe. In this setting, the importance and utility of Early Warning Systems is becoming increasingly obvious. In this work, we present the design of an early warning system called TransCPEarlyWarning, aimed at seven countries in the Adriatic-Ionian area in Europe. The overall objective is to increase the level of cooperation among national civil protection institutions in these countries, addressing natural and man-made risks from the early warning stage and improving the intervention capabilities of civil protection mechanisms. The system utilizes an innovative approach with a lever effect, while also aiming to support the whole system of Civil Protection.

cs.CY↗

arXiv2021

Local Learning at the Network Edge for Efficient & Secure Real-Time Predictive Analytics

The ability to perform computation on devices, such as smartphones, cars, or other nodes present at the Internet of Things leads to constraints regarding bandwidth, storage, and energy, as most of these devices are mobile and operate on batteries. Using their computational power to perform locally machine learning and analytics tasks can enable accurate and real-time predictions at the network edge. A trained machine learning model requires high accuracy towards the prediction outcome, as wrong decisions can lead to negative consequences on the efficient conclusion of applications. Most of the data sensed in these devices are contextual and personal requiring privacy-preserving without their distribution over the network. When working with these privacy-preserving data, not only the protection is important but, also, the model needs the ability to adapt to regular occurring concept drifts and data distribution changes to guarantee a high accuracy of the prediction outcome. We address the importance of personalization and generalization in edge devices to adapt to data distribution updates over continuously evolving environments. The methodology we propose relies on the principles of Federated Learning and Optimal Stopping Theory extended with a personalization component. The privacy-efficient and quality-awareness of personalization and generalization is the overarching aim of this work.

cs.DC↗

arXiv2021

Max-Utility Based Arm Selection Strategy For Sequential Query Recommendations

We consider the query recommendation problem in closed loop interactive learning settings like online information gathering and exploratory analytics. The problem can be naturally modelled using the Multi-Armed Bandits (MAB) framework with countably many arms. The standard MAB algorithms for countably many arms begin with selecting a random set of candidate arms and then applying standard MAB algorithms, e.g., UCB, on this candidate set downstream. We show that such a selection strategy often results in higher cumulative regret and to this end, we propose a selection strategy based on the maximum utility of the arms. We show that in tasks like online information gathering, where sequential query recommendations are employed, the sequences of queries are correlated and the number of potentially optimal queries can be reduced to a manageable size by selecting queries with maximum utility with respect to the currently executing query. Our experimental results using a recent real online literature discovery service log file demonstrate that the proposed arm selection strategy improves the cumulative regret substantially with respect to the state-of-the-art baseline algorithms. % and commonly used random selection strategy for a variety of contextual multi-armed bandit algorithms. Our data model and source code are available at ~\url{https://anonymous.4open.science/r/0e5ad6b7-ac02-4577-9212-c9d505d3dbdb/}.

cs.LG↗

arXiv2021

A Proactive Management Scheme for Data Synopses at the Edge

The combination of the infrastructure provided by the Internet of Things (IoT) with numerous processing nodes present at the Edge Computing (EC) ecosystem opens up new pathways to support intelligent applications. Such applications can be provided upon humongous volumes of data collected by IoT devices being transferred to the edge nodes through the network. Various processing activities can be performed on the discussed data and multiple collaborative opportunities between EC nodes can facilitate the execution of the desired tasks. In order to support an effective interaction between edge nodes, the knowledge about the geographically distributed data should be shared. Obviously, the migration of large amounts of data will harm the stability of the network stability and its performance. In this paper, we recommend the exchange of data synopses than real data between EC nodes to provide them with the necessary knowledge about peer nodes owning similar data. This knowledge can be valuable when considering decisions such as data/service migration and tasks offloading. We describe an continuous reasoning model that builds a temporal similarity map of the available datasets to get nodes understanding the evolution of data in their peers. We support the proposed decision making mechanism through an intelligent similarity extraction scheme based on an unsupervised machine learning model, and, at the same time, combine it with a statistical measure that represents the trend of the so-called discrepancy quantum. Our model can reveal the differences in the exchanged synopses and provide a datasets similarity map which becomes the appropriate knowledge base to support the desired processing activities. We present the problem under consideration and suggest a solution for that, while, at the same time, we reveal its advantages and disadvantages through a large number of experiments.

cs.LG↗

arXiv2021

Cooperative Multi-Modal Localization in Connected and Autonomous Vehicles

Cooperative Localization is expected to play a crucial role in various applications in the field of Connected and Autonomous vehicles (CAVs). Future 5G wireless systems are expected to enable cost-effective Vehicle-to-Everything (V2X)systems, allowing CAVs to share with the other entities of the network the data they collect and measure. Typical measurement models usually deployed for this problem, are absolute position from Global Positioning System (GPS), relative distance and azimuth angle to neighbouring vehicles, extracted from Light Detection and Ranging (LIDAR) or Radio Detection and Ranging(RADAR) sensors. In this paper, we provide a cooperative localization approach that performs multi modal-fusion between the interconnected vehicles, by representing a fleet of connected cars as an undirected graph, encoding each vehicle position relative to its neighbouring vehicles. This method is based on:i) the Laplacian Processing, a Graph Signal Processing tool that allows to capture intrinsic geometry of the undirected graph of vehicles rather than their absolute position on global coordinate system and ii) the temporal coherence due to motion patterns of the moving vehicles.

eess.SP↗