Periodic Reporting for period 1 - TRUMPET (TRUstworthy Multi-site Privacy Enhancing Technologies)
Periodo di rendicontazione: 2022-10-01 al 2024-03-31
Federated Learning (FL) has gained prominence as a privacy-enhancing technology, initially popularized by Google for predicting user keystrokes on smartphones while preserving privacy. FL has expanded to various applications and novel algorithms have been developed to ensure the convergence of FL's global AI model and handle intermittent connectivity in certain scenarios.
However, the strength of privacy protection provided by FL has been called into question by further research. Researchers have identified potential vulnerabilities and threats, such as the curious aggregator threat, susceptibility to man-in-the-middle and insider attacks that disrupt model convergence, and inference attacks that attempt to re-identify data subjects from AI model updates. To address these challenges, the novel Armored Federated Learning (AFL) platform will be developed in this project, aiming to meet the GDPR challenge and enable GDPR compliance while utilizing FL technology.
Another challenge is the accessibility of high-quality datasets for research purposes. Often, datasets of sufficient quality are not available from a single source and must be assembled from subsets owned by different organizations, each with its own access policies. In many cases, these organizations restrict access to entities outside their organization, even more so than required by the GDPR. Furthermore, the mere awareness of the dataset's existence and the ability to contact the data owner pose additional challenges to data accessibility. Overcoming the barrier of accessibility to siloed data in a GDPR-compliant manner would have a significant impact.
Overall, FL and improved accessibility to siloed data in a GDPR-compliant manner hold the potential to revolutionize privacy protection, research capabilities, and the development of critical solutions in various domains.
The goal of TRUMPET project is to research and develop novel privacy enhancement methods for Federated Learning, and to deliver a highly scalable Federated AI service platform for researchers, that will enable AI-powered studies of siloed, multi-site, cross-domain, cross-border European datasets with privacy guarantees that exceed the requirements of GDPR. The generic TRUMPET platform will be piloted, demonstrated and validated in the specific use case of European cancer hospitals, allowing researchers and policymakers to extract AI-driven insights from previously inaccessible cross-border, cross-organization cancer data, while ensuring the patients’ privacy. The strong privacy protection accorded by the platform will be verified through the engagement of external experts for independent privacy leakage and re-identification testing.
A secondary goal of the project is to research, develop and promote with EU data protection authorities a novel metric and tool for the certification of GDPR compliance of Federated Learning implementations.
However, the strength of privacy protection provided by FL has been called into question by further research. Researchers have identified potential vulnerabilities and threats, such as the curious aggregator threat, susceptibility to man-in-the-middle and insider attacks that disrupt model convergence, and inference attacks that attempt to re-identify data subjects from AI model updates. To address these challenges, the novel Armored Federated Learning (AFL) platform will be developed in this project, aiming to meet the GDPR challenge and enable GDPR compliance while utilizing FL technology.
Another challenge is the accessibility of high-quality datasets for research purposes. Often, datasets of sufficient quality are not available from a single source and must be assembled from subsets owned by different organizations, each with its own access policies. In many cases, these organizations restrict access to entities outside their organization, even more so than required by the GDPR. Furthermore, the mere awareness of the dataset's existence and the ability to contact the data owner pose additional challenges to data accessibility. Overcoming the barrier of accessibility to siloed data in a GDPR-compliant manner would have a significant impact.
Overall, FL and improved accessibility to siloed data in a GDPR-compliant manner hold the potential to revolutionize privacy protection, research capabilities, and the development of critical solutions in various domains.
The goal of TRUMPET project is to research and develop novel privacy enhancement methods for Federated Learning, and to deliver a highly scalable Federated AI service platform for researchers, that will enable AI-powered studies of siloed, multi-site, cross-domain, cross-border European datasets with privacy guarantees that exceed the requirements of GDPR. The generic TRUMPET platform will be piloted, demonstrated and validated in the specific use case of European cancer hospitals, allowing researchers and policymakers to extract AI-driven insights from previously inaccessible cross-border, cross-organization cancer data, while ensuring the patients’ privacy. The strong privacy protection accorded by the platform will be verified through the engagement of external experts for independent privacy leakage and re-identification testing.
A secondary goal of the project is to research, develop and promote with EU data protection authorities a novel metric and tool for the certification of GDPR compliance of Federated Learning implementations.
The TRUMPET project has made significant advancements in Federated Learning (FL) methodologies while ensuring compliance with legal, ethical, and regulatory standards. Initially, an exhaustive survey identified privacy threats and risks in FL, leading to comprehensive system requirements and the development of intuitive dashboards and detailed architectural plans for the TRUMPET platform. Key deliverables like D1.3 provided detailed software specifications, facilitating subsequent development tasks. The project also evaluated various Privacy-Enhancing Technologies (PETs), such as Homomorphic Encryption and Secure Multi-Party Computation, prioritizing the most suitable combinations and implementing innovative techniques like Lagrange Coded Computing. Efforts in Deliverable D2.2 focused on refining and benchmarking PET methods. Additionally, the project examined threat models, adversary assumptions, and privacy metrics, with practical evaluations documented in deliverables like D3.1 and D3.2. Concurrently, data acquisition and integration were advanced through a microservices architecture and Monorepo project structure, ensuring data integrity and compliance. Collaborative efforts established data format standards, furthering the development of the FL platform and PET implementation frameworks, highlighting the project's commitment to continuous innovation and refinement.
Federated Learning is an emerging technology that allows to exploit large amounts of data distributed across multiple institutions without exposing their private datasets. However, despite its high potential for creating a positive impact, existing Federated Learning implementations still remain with a lower adoption due, mainly, to the unresolved privacy issues that increase the risk to comply with the GDPR requirements.
The TRUMPET project is strongly focused on tackling these privacy issues, aiming first at building a highly scalable, cross-border federated and privacy-enhanced platform dedicated to AI-based studies that will assist healthcare professionals. With this main objective in mind, the project will deal with two complementary and more technical aspects: (1) develop novel privacy preserving technologies tailored for the scenario of Federated Learning, (2) develop a novel privacy measurement tool to assess the privacy risks in the TRUMPET platform, and in other existing Federated Learning implementations. Last, but no least, the project will identify and perform a legal study of the implications of GDPR in Federated Learning implementations.
Building on these results, our platform will enable AI-powered studies of cross-border European datasets with privacy guarantees that follow the requirements of GDPR. TRUMPET research and innovation activities will aid to support digital innovations while highly preserving privacy, security and ethical standards. Consequently, TRUMPET will increase the adoption of Federated Learning systems among data owners and researchers and, thereby, it will contribute to process medical data collaboratively to find more accurate explanations for diseases and to reduce health risks.
The TRUMPET project is strongly focused on tackling these privacy issues, aiming first at building a highly scalable, cross-border federated and privacy-enhanced platform dedicated to AI-based studies that will assist healthcare professionals. With this main objective in mind, the project will deal with two complementary and more technical aspects: (1) develop novel privacy preserving technologies tailored for the scenario of Federated Learning, (2) develop a novel privacy measurement tool to assess the privacy risks in the TRUMPET platform, and in other existing Federated Learning implementations. Last, but no least, the project will identify and perform a legal study of the implications of GDPR in Federated Learning implementations.
Building on these results, our platform will enable AI-powered studies of cross-border European datasets with privacy guarantees that follow the requirements of GDPR. TRUMPET research and innovation activities will aid to support digital innovations while highly preserving privacy, security and ethical standards. Consequently, TRUMPET will increase the adoption of Federated Learning systems among data owners and researchers and, thereby, it will contribute to process medical data collaboratively to find more accurate explanations for diseases and to reduce health risks.