Security and Privacy Accountable Technology Innovations, Algorithms, and machine Learning

Resultado final

Based on task 41 this deliverable provides a framework for determining key processes and people to enable holistic technology development within this project with emphasis on the social legal and technical limitations and gender and diversity considerations

Kick-off Meeting Report

A summary of kickmeeting of SPATIALT71

Requirements Analysis for AI towards Addressing Security Risks and Threats to System and Network Architectures

This deliverable is based on outcomes of Task 11 as well as Task 15 and will incorporate an initial yet extensive requirements analysis that facilitates AI enabled and secure system architectures

Impact Assessment and Exploitation Interim Report

Report of indicators measures and elements designed and released in SPATIAL Impact Master Plan through the first period of the project T63 and T64

Collated internal advice on SPATIAL project

Final report covering end results(T7.4)

Final Requirements Analysis for AI towards Addressing Security Risks and Threats to System and Network Architectures

This deliverable is based on outcomes of Task 11 as well as Task 15 and will incorporate the final indepth requirements analysis that enables AI enabled and secure system architectures

An explanatory platform that accounts AI systems based on its quantified quality

This deliverable is based on the outcome of Task 35 which will also contribute to WP4 and WP5 It describes a novel mechanism to leverage Trusted Execution Environments towards making AI accountable This novel mechanism is also scalable with a massive amount of data under strict timing constraints for AI applications

Initial technology transfer and recommendations derived from each pilot

Tasks involved T51T54 Deliverable will record initial status of all the pilots from the first round completed lessons learned from the integration and experiments performed and provide initial technology transfer and recommendations to AI component design partners from WP1 WP2 and WP3 Also demonstrators will be provided per usecase video live demo executable etc

Automated diagnosis and mechanisms for tuning AI-based systems and Trusted Execution Environments for accountable and resilient AI

This deliverable is based on the outcome of Task 35 which will also contribute to WP4 and WP5 It describes a novel mechanism to leverage Trusted Execution Environments towards making AI accountable This novel mechanism is also scalable with a massive amount of data under strict timing constraints for AI applications

Final technology transfer and recommendations derived from each pilot

Tasks involved: T5.1-T5.4. Deliverable will describe last improvements and validations made in the pilots and will include the final version of the components (depends on type: open source, executables, commercial) and demonstrators to show how the components are used per use case (via video, live demo, etc.). It will analyse the effective usability (i.e., KPIs) of the SPATIAL results in the pilots, and provide recommendations and identify future developments for the industrialisation of SPATIAL's results that will serve to define the exploitation plans in D6.5.

Quality Assurance Plan

Composition of quality T73

Process to Integrate Accountability and Resilience Features into AI Algorithms

This deliverable is based on the outcome of Task 23 It describes a general process proposed to make existing AI algorithms accountable and resilient The metrics proposed in the deliverable 22 will be used to evaluate the performance of this process

Detection mechanisms to identify data biases and exploratory studies about different data quality trade-offs for AI-based systems

This deliverable is based on the outcomes of Task 31 and Task 32 It presents an analysis of different AI model behaviors that are obtained when machine learning is trained over distributed and federated infrastructure

Project Management Plan

General management framework for SPATIALT72

Dissemination and Communication Final Report

Report of indicators, measures and elements designed and released to run the dissemination and communication plans through the second period of the project (T6.1 and T6.2).

Security Threats modelling for AI based System Architectures

This deliverable is based on outcomes of Tasks 11 and 12 and will incorporate the results of analysis to identify potential threats to AI based systems and to derive security requirements for countering the identified threats

Performance evaluation in controlled environments and guidelines to build the pilot studies in real testbeds

This deliverable is based on outcomes of Task 3.6 and will incorporate the performance evaluation of the envisioned platform in different use cases and environments.

Field research analysis report and integration action plan

A publicly available open access and peerreviewed scientific research article which will serve to define process development practices based on tasks 41 and 42

Research Data Management Plan

Summary of open access and data management procedure in SPATIALT71

Impact Assessment and Exploitation Final Report

Report of indicators, measures and elements designed and released in SPATIAL Impact Master Plan through the first period of the project (T6.3 and T6.4).

Initial description of the use-cases, design, testbed, experimentation for all the pilots

Tasks involved T51T54 Deliverable will set the scene for all the pilots experimental environments including description of the use cases testbed and data used for each and design of experiments per usecase

Education module full launch

This deliverable is based on task 4.5, consisting of an educational module focusing on socio-technical skills and ethical socio-legal awareness for current and future AI engineers and developers to ensure the development of accountable security solutions.

Design Principles for Accountable and Resilient AI Architectures

This deliverable is based on outcomes of Task 1.1, Task 1.3 as well as Task 1.5 and will incorporate derived design principles for accountable and resilient AI Architectures based on the conducted requirements analyses and non-functional aspects

Sociotechnical, regulatory and ethical implications and integration guidelines report

This deliverable emerges from tasks 4.3 and 4.4. This report will incorporate information on how SPATIAL development of explainability, transparency and overall “algorithm traceability” can mitigate some of the regulatory and ethical issues and ensure greater transparency. It will include guidelines for the use of SPATIAL techniques and tools as well as best practices.

Robust Accountability Metrics for AI Algorithms

This deliverable will provide metrics from T22 to assess 1 quality and usefulness of explanation of AI models 2 tradeoff between AI model utility of and privacy of usersdataowners 3 resilience of models built to ML attacks

Dissemination & communication interim report

Report of indicators measures and elements designed and released to run the dissemination and communication plans through the first period of the project T61 and T62

Existing AI Algorithms and their Accountability and Resilience Features within the Context of Applications to IoT, 5G, and Cybersecurity

This deliverable is based on outcomes of Task 21 and will incorporate an analysis of existing AI algorithms and their characteristics related to accountability as well as resilience in the context of applications to IoT 5G and cyber security

SPATIAL Impact Master Plan

Master plan for the dissemination communication and exploitation strategies T62 T63 designed for the project The plan will also provide an overview of the stakeholder base to target T61 as well as a standardization landscape as the basis for further standardization activities in the project T64

Publicaciones

Privacy-preserving AI for future networks

Autores: Diego Perino , Kleomenis Katevas , Andra Lutu , Eduard Marin , Nicolas Kourtellis
Publicado en: Communications of the ACM Volume 65 Edición 4April 2022, 2022, ISSN 0001-0782
Editor: Association for Computing Machinary, Inc.
DOI: 10.1145/3512343

The Many Faces of Edge Intelligence

Autores: Ella Peltonen; Ijaz Ahmad; Atakan Aral; Michele Capobianco; Aaron Yi Ding; Felipe Gil-Castiñeira; Ekaterina Gilman; Erkki Harjula; Marko Jurmu; Teemu Karvonen; Markus Kelanti; Teemu Leppänen; Lauri Lovén; Tommi Mikkonen; Nitinder Mohan; Petteri Nurmi; Susanna Pirttikangas; Paweł Sroka; Sasu Tarkoma; Tingting Yang
Publicado en: IEEE Access ( Volume: 10), 2022, ISSN 2169-3536
Editor: Institute of Electrical and Electronics Engineers Inc.
DOI: 10.1109/access.2022.3210584

A Survey on Privacy of Personal and Non-Personal Data in B5G/6G Networks

Autores: Chamara Sandeepa, Bartlomiej Siniarski, Nicolas Kourtellis, Shen Wang, Madhusanka Liyanage
Publicado en: ACM Computing Surveys, Edición 56, 2024, Página(s) 1-37, ISSN 0360-0300
Editor: Association for Computing Machinary, Inc.
DOI: 10.1145/3662179

Social-aware Federated Learning: Challenges and Opportunities in Collaborative Data Training

Autores: Ottun, Abdul-Rasheed, Pramod C. Mane, Zhigang Yin, Souvik Paul, Mohan Liyanage, Jason Pridmore, Aaron Yi Ding, Rajesh Sharma, Petteri Nurmi, and Huber Flores
Publicado en: IEEE Internet Computing Magazine, 2022, ISSN 1089-7801
Editor: Institute of Electrical and Electronics Engineers
DOI: 10.1109/mic.2022.3219263

Opportunistic Multi-Drone Networks: Filling the Spatiotemporal Holes of Collaborative and Distributed Applications

Autores: Huber Flores
Publicado en: IEEE Internet of Things Magazine, Edición 7, 2024, Página(s) 94-100, ISSN 2576-3180
Editor: IEEE Internet of Things Magazine, vol. 7, no. 2, pp. 94-100
DOI: 10.1109/iotm.001.2300169

A survey on privacy for B5G/6G: New privacy challenges, and research directions

Autores: Chamara Sandeepa, Bartlomiej Siniarski, Nicolas Kourtellis, Shen Wang, Madhusanka Liyanage
Publicado en: Journal of Industrial Information Integration, 2022, ISSN 2452-414X
Editor: ELSEVIER
DOI: 10.1016/j.jii.2022.100405

Dynamic-IMD (D-IMD): Introducing activity spaces to deprivation measurement in London, Birmingham and Liverpool

Autores: Sam Comber, Souneil Park , Daniel Arribas-Bel
Publicado en: Cities Volume 127, August 2022, 103733, 2022, ISSN 0264-2751
Editor: Pergamon Press Ltd.
DOI: 10.1016/j.cities.2022.103733

A Survey on Approximate Edge AI for Energy Efficient Autonomous Driving Services

Autores: Dewant Katare, Diego Perino, Jari Nurmi, Martijn Warnier, Marijn Janssen, Aaron Yi Ding
Publicado en: IEEE Communications Surveys & Tutorials, Edición 25, 2024, Página(s) 2714-2754, ISSN 1553-877X
Editor: Institute of Electrical and Electronics Engineers
DOI: 10.1109/comst.2023.3302474

Tiny, Always-on, and Fragile: Bias Propagation through Design Choices in On-device Machine Learning Workflows

Autores: Wiebke (Toussaint) Hutiri, Aaron Yi Ding, Fahim Kawsar, Akhil Mathur
Publicado en: ACM Transactions on Software Engineering and Methodology, Edición 32, 2023, Página(s) 1-37, ISSN 1049-331X
Editor: Association for Computing Machinary, Inc.
DOI: 10.1145/3591867

Roadmap for edge AI: a Dagstuhl perspective

Autores: Aaron Yi Ding , Ella Peltonen , Tobias Meuser , Atakan Aral , Christian Becker , Schahram Dustdar , Thomas Hiessl , Dieter Kranzlmüller , Madhusanka Liyanage , Setareh Maghsudi , Nitinder Mohan , Jörg Ott , Jan S. Rellermeyer , Stefan Schulte , Henning Schulzrinne , Gürkan Solmaz , Sasu Tarkoma , Blesson Varghese , Lars Wolf
Publicado en: ACM SIGCOMM Computer Communication Review, Edición Vol. 52, No. 1, 2022, ISSN 0146-4833
Editor: Association for Computing Machinery
DOI: 10.1145/3523230.3523235

AI Sensors and Dashboards

Autores: Huber Flores
Publicado en: Computer, Edición 57, 2024, Página(s) 55-64, ISSN 0018-9162
Editor: Institute of Electrical and Electronics Engineers
DOI: 10.1109/mc.2024.3394056

Rethinking Artificial Intelligence: Algorithmic Bias and Ethical Issues| Rage Against the Artificial Intelligence? Understanding Contextuality of Algorithm Aversion and Appreciation

Autores: OOMEN, Tessa; GONÇALVES, João; MOLS, Anouk.
Publicado en: International Journal of Communication, 2024, ISSN 1932-8036
Editor: USC Annenberg School for Communication & Journalism
DOI: 10.5281/zenodo.10986943

Trustworthy and Sustainable Edge AI: A Research Agenda

Autores: Aaron Yi Ding; Marijn Janssen; Jon Crowcroft
Publicado en: 2021 Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA), 2022
Editor: IEEE
DOI: 10.1109/tpsisa52974.2021.00019

Exploring CNN and XAI-based Approaches for Accountable MI Detection in the Context of IoT-enabled Emergency Communication Systems

Autores: Helene Knof, Prachi Bagave, Michell Boerger, Nikolay Tcholtchev, Aaron Yi Ding
Publicado en: Proceedings of the International Conference on the Internet of Things, Edición 22, 2024, Página(s) 50-57
Editor: ACM
DOI: 10.1145/3627050.3627057

Who Funds Misinformation? A Systematic Analysis of the Ad-related Profit Routines of Fake News Sites

Autores: Emmanouil Papadogiannakis, Panagiotis Papadopoulos, Evangelos P. Markatos, Nicolas Kourtellis
Publicado en: Proceedings of the ACM Web Conference 2023, 2023, Página(s) 2765-2776
Editor: ACM
DOI: 10.1145/3543507.3583443

Bias in Automated Speaker Recognition

Autores: Wiebke Toussaint Hutiri , Aaron Yi Ding
Publicado en: FAccT '22: 2022 ACM Conference on Fairness, Accountability, and Transparency, 2022, ISBN 978-1-4503-9352-2
Editor: Association for Computing Machinery
DOI: 10.1145/3531146.3533089

A Large-scale Examination of ”Socioeconomic” Fairness in Mobile Networks

Autores: Souneil Park , Pavol Mulinka , Diego Perino
Publicado en: COMPASS '22: ACM SIGCAS/SIGCHI Conference on Computing and Sustainable Societies (COMPASS), 2022, ISBN 978-1-4503-9347-8
Editor: COMPASS '22: ACM SIGCAS/SIGCHI Conference on Computing and Sustainable Societies
DOI: 10.1145/3530190.3534809

FLAME: Taming Backdoors in Federated Learning

Autores: Thien Duc Nguyen, Phillip Rieger, Huili Chen, Hossein Yalame, Helen Möllering, Hossein Fereidooni, Samuel Marchal, Markus Miettinen, Azalia Mirhoseini, Shaza Zeitouni, Farinaz Koushanfar, Ahmad-Reza Sadeghi, Thomas Schneider
Publicado en: 2022
Editor: 31st USENIX Security Symposium, August 2022, Boston, MA, USA
DOI: 10.48550/arxiv.2101.02281

Not only E.T. Phones Home: Analysing the Native User Tracking of Mobile Browsers

Autores: John Pegioudis, Emmanouil Papadogiannakis, Nicolas Kourtellis, Evangelos P. Markatos, Panagiotis Papadopoulos
Publicado en: Proceedings of the 2023 ACM on Internet Measurement Conference, 2024, Página(s) 181-187
Editor: ACM
DOI: 10.1145/3618257.3624842

SPATIAL: Practical AI Trustworthiness with Human Oversight

Autores: Abdul-Rasheed Ottun, Rasinthe Marasinghe, Toluwani Elemosho, Mohan Liyanage, Ashfaq Hussain Ahmed, Michell Boerger, Chamara Sandeepa, Thulitha Senevirathna, Vinh Hoa La, Manh-Dung Nguyen, Claudio Soriente, Samuel Marchal, Shen Wang, David Solans Noguero, Nikolay Tcholtchev, Aaron Yi Ding, Huber Flores
Publicado en: 2024 IEEE 44th International Conference on Distributed Computing Systems (ICDCS), 2024, Página(s) 1427-1430
Editor: IEEE
DOI: 10.1109/icdcs60910.2024.00138

Gender Disparities in Child Custody Sentencing in Spain

Autores: Júlia Riera, David Solans, Marzieh Karimi-Haghighi, Carlos Castillo, Caterina Calsamiglia
Publicado en: Proceedings of the Nineteenth International Conference on Artificial Intelligence and Law, Edición 81, 2024, Página(s) 237-246
Editor: ACM
DOI: 10.1145/3594536.3595135

Towards Accountable and Resilient AI-Assisted Networks: Case Studies and Future Challenges

Autores: Shen Wang, Chamara Sandeepa, Thulitha Senevirathna, Bartlomiej Siniarski, Manh-Dung Nguyen, Samuel Marchal, Madhusanka Liyanage
Publicado en: 2024 Joint European Conference on Networks and Communications &amp; 6G Summit (EuCNC/6G Summit), 2024, Página(s) 818-823
Editor: IEEE
DOI: 10.1109/eucnc/6gsummit60053.2024.10597060

Robust and Resilient Federated Learning for Securing Future Networks

Autores: Yushan Siriwardhana; Pawani Porambage; Madhusanka Liyanage; Mika Ylianttila
Publicado en: 2022 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit), 2022
Editor: IEEE
DOI: 10.1109/eucnc/6gsummit54941.2022.9815812

SHERPA: Explainable Robust Algorithms for Privacy-Preserved Federated Learning in Future Networks to Defend Against Data Poisoning Attacks

Autores: Chamara Sandeepa, Bartlomiej Siniarski, Shen Wang, Madhusanka Liyanage
Publicado en: 2024 IEEE Symposium on Security and Privacy (SP), Edición 34, 2024, Página(s) 4772-4790
Editor: IEEE
DOI: 10.1109/sp54263.2024.00271

The Hitchhiker’s Guide to Facebook Web Tracking with Invisible Pixels and Click IDs

Autores: Paschalis Bekos, Panagiotis Papadopoulos, Evangelos P. Markatos, Nicolas Kourtellis
Publicado en: Proceedings of the ACM Web Conference 2023, 2024
Editor: ACM
DOI: 10.1145/3543507.3583311

Accountable AI for Healthcare IoT Systems

Autores: Prachi Bagave, Marcus Westberg, Roel Dobbe, Marijn Janssen, Aaron Yi Ding
Publicado en: 2022 IEEE 4th International Conference on Trust, Privacy and Security in Intelligent Systems, and Applications (TPS-ISA), Edición 55, 2024, Página(s) 20-28
Editor: IEEE
DOI: 10.1109/tps-isa56441.2022.00013

Federated Learning based Anomaly Detection as an Enabler for Securing Network and Service Management Automation in Beyond 5G Networks

Autores: Suwani Jayasinghe Centre for Wireless Communications, University of Oulu, Finland ; Yushan Siriwardhana; Pawani Porambage; Madhusanka Liyanage; Mika Ylianttila
Publicado en: 2022 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit), 2022
Editor: ieee
DOI: 10.1109/eucnc/6gsummit54941.2022.9815754

Towards Trustworthy Edge Intelligence: Insights from Voice-Activated Services

Autores: Wiebke Toussaint Hutiri; Aaron Yi Ding
Publicado en: 2022 IEEE International Conference on Services Computing (SCC), 2022
Editor: IEEE
DOI: 10.1109/scc55611.2022.00043

The SPATIAL Architecture: Design and Development Experiences from Gauging and Monitoring the AI Inference Capabilities of Modern Applications

Autores: Abdul-Rasheed Ottun, Rasinthe Marasinghe, Toluwani Elemosho, Mohan Liyanage, Mohamad Ragab, Prachi Bagave, Marcus Westberg, Mehrdad Asadi, Michell Boerger, Chamara Sandeepa, Thulitha Senevirathna, Bartlomiej Siniarski, Madhusanka Liyanage, Vinh Hoa La, Manh-Dung Nguyen, Edgardo Montes De Oca, Tessa Oomen, João Fernando Ferreira Gonçalves, Illija Tanasković, Sasa Klopanovic, Nicolas Kourtellis,
Publicado en: 2024 IEEE 44th International Conference on Distributed Computing Systems (ICDCS), Edición 31, 2024, Página(s) 947-959
Editor: IEEE
DOI: 10.1109/icdcs60910.2024.00092

One to Rule them All: A Study on Requirement Management Tools for the Development of Modern AI-based Software

Autores: Abdul-Rasheed Ottun, Mehrdad Asadi, Michell Boerger, Nikolay Tcholtchev, João Gonçalves, Duşan Borovčanin, Bartlomiej Siniarsk, Huber Flores
Publicado en: 2023 IEEE International Conference on Big Data (BigData), Edición 64, 2024, Página(s) 3556-3565
Editor: IEEE
DOI: 10.1109/bigdata59044.2023.10386926

Privacy–Preserving Online Content Moderation: A Federated Learning Use Case

Autores: Pantelitsa Leonidou, Nicolas Kourtellis, Nikos Salamanos, Michael Sirivianos
Publicado en: Companion Proceedings of the ACM Web Conference 2023, Edición 9, 2024, Página(s) 280-289
Editor: ACM
DOI: 10.1145/3543873.3587604

Characterising the Role of Pre-Processing Parameters in Audio-based Embedded Machine Learning

Autores: Wiebke Toussaint , Akhil Mathur , Aaron Yi Ding , Fahim Kawsar
Publicado en: SenSys '21: Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems, 2021, ISBN 978-1-4503-9097-2
Editor: Association for Computing Machinery
DOI: 10.1145/3485730.3493448

A longitudinal study of the top 1% toxic Twitter profiles

Autores: Hina Qayyum, Benjamin Zi Hao Zhao, Ian Wood, Muhammad Ikram, Nicolas Kourtellis, Mohamad Ali Kaafar
Publicado en: Proceedings of the 15th ACM Web Science Conference 2023, 2024, Página(s) 292-303
Editor: ACM
DOI: 10.1145/3578503.3583619

A deep learning anomaly detection framework with explainability and robustness

Autores: Manh-Dung Nguyen, Anis Bouaziz, Valeria Valdes, Ana Rosa Cavalli, Wissam Mallouli, Edgardo Montes De Oca
Publicado en: Proceedings of the 18th International Conference on Availability, Reliability and Security, Edición 201, 2024, Página(s) 1-7
Editor: ACM
DOI: 10.1145/3600160.3605052

Subversion-Resilient Enhanced Privacy ID

Autores: Antonio Faonio, Dario Fiore, Luca Nizzardo & Claudio Soriente
Publicado en: Lecture Notes in Computer Science, 2022, ISBN 978-3-030-95311-9
Editor: Springer, Cham
DOI: 10.1007/978-3-030-95312-6_23

No Forking Way: Detecting Cloning Attacks on Intel SGX Applications

Autores: Samira Briongos, Ghassan Karame, Claudio Soriente, Annika Wilde
Publicado en: Annual Computer Security Applications Conference, Edición 2015, 2023, Página(s) 744-758
Editor: ACM
DOI: 10.1145/3627106.3627187

Quantitative Evaluation of xAI Methods for Multivariate Time Series - A Case Study for a CNN-Based MI Detection Model

Autores: Helene Knof, Michell Boerger, Nikolay Tcholtchev
Publicado en: Communications in Computer and Information Science, Explainable Artificial Intelligence, 2024, Página(s) 169-190
Editor: Springer Nature Switzerland
DOI: 10.1007/978-3-031-63803-9_9

Resilience of Blockchain Overlay Networks

Autores: Aristodemos Paphitis, Nicolas Kourtellis, Michael Sirivianos
Publicado en: Lecture Notes in Computer Science, Network and System Security, 2023, Página(s) 93-113
Editor: Springer Nature Switzerland
DOI: 10.1007/978-3-031-39828-5_6

From Opacity to Clarity: Leveraging XAI for Robust Network Traffic Classification

Autores: Chamara Sandeepa, Thulitha Senevirathna, Bartlomiej Siniarski, Manh-Dung Nguyen, Vinh-Hoa La, Shen Wang, Madhusanka Liyanage
Publicado en: Communications in Computer and Information Science, Asia Pacific Advanced Network, 2024, Página(s) 125-138
Editor: Springer Nature Switzerland
DOI: 10.1007/978-3-031-51135-6_11

Graph Analysis of Blockchain P2P Overlays and Their Security Implications

Autores: Aristodemos Paphitis, Nicolas Kourtellis, Michael Sirivianos
Publicado en: Lecture Notes in Computer Science, Security and Privacy in Social Networks and Big Data, 2023, Página(s) 167-186
Editor: Springer Nature Singapore
DOI: 10.1007/978-981-99-5177-2_10

Toward Anomaly Detection Using Explainable AI

Autores: Manh-Dung Nguyen, Vinh-Hoa La, Wissam Mallouli, Ana Rosa Cavalli, Edgardo Montes de Oca
Publicado en: CyberSecurity in a DevOps Environment, 2023, Página(s) 293-324
Editor: Springer Nature Switzerland
DOI: 10.1007/978-3-031-42212-6_10

Derechos de propiedad intelectual

Montimage Monitoring Tool (MMT) framework

Número de solicitud/publicación: MMT 2
Fecha: 2023-01-01
Solicitante(s): MONTIMAGE EURL

Montimage AI Platform (MAIP)

Número de solicitud/publicación: MMT 1
Fecha: 2023-01-01
Solicitante(s): MONTIMAGE EURL

Resultado final

Publicaciones

Derechos de propiedad intelectual

Compartir esta página Compartir esta página en las redes sociales

Descargar Descargar el contenido de la página