Projects per year
Abstract
Objectives
To assess a range of tools and methods to support Trusted Research Environments (TREs) to assess output from AI methods for potentially identifiable information, investigate the legal and ethical implications and controls, and produce a set of guidelines and recommendations to support all TREs with export controls of AI algorithms.
Approach
TREs provide secure facilities to analyse confidential personal data, with staff checking outputs for disclosure risk before publication. Artificial intelligence (AI) has high potential to improve the linking and analysis of population data, and TREs are well suited to supporting AI modelling. However, TRE governance focuses on classical statistical data analysis. The size and complexity of AI models presents significant challenges for the disclosure-checking process. Models may be susceptible to external hacking: complicated methods to reverse engineer the learning process to find out about the data used for training, with more potential to lead to re-identification than conventional statistical methods.
Results
GRAIMatter is:
Quantitatively assessing the risk of disclosure from different AI models exploring different models, hyper-parameter settings and training algorithms over common data types
Evaluating a range of tools to determine effectiveness for disclosure control
Assessing the legal and ethical implications of TREs supporting AI development and identifying aspects of existing legal and regulatory frameworks requiring reform.
Running 4 PPIE workshops to understand their priorities and beliefs around safeguarding and securing data
Developing a set of recommendations including
suggested open-source toolsets for TREs to use to measure and reduce disclosure risk
descriptions of the technical and legal controls and policies TREs should implement across the 5 Safes to support AI algorithm disclosure control
training implications for both TRE staff and how they validate researchers
Conclusion
GRAIMatter is developing a set of usable recommendations for TREs to use to guard against the additional risks when disclosing trained AI models from TREs.
To assess a range of tools and methods to support Trusted Research Environments (TREs) to assess output from AI methods for potentially identifiable information, investigate the legal and ethical implications and controls, and produce a set of guidelines and recommendations to support all TREs with export controls of AI algorithms.
Approach
TREs provide secure facilities to analyse confidential personal data, with staff checking outputs for disclosure risk before publication. Artificial intelligence (AI) has high potential to improve the linking and analysis of population data, and TREs are well suited to supporting AI modelling. However, TRE governance focuses on classical statistical data analysis. The size and complexity of AI models presents significant challenges for the disclosure-checking process. Models may be susceptible to external hacking: complicated methods to reverse engineer the learning process to find out about the data used for training, with more potential to lead to re-identification than conventional statistical methods.
Results
GRAIMatter is:
Quantitatively assessing the risk of disclosure from different AI models exploring different models, hyper-parameter settings and training algorithms over common data types
Evaluating a range of tools to determine effectiveness for disclosure control
Assessing the legal and ethical implications of TREs supporting AI development and identifying aspects of existing legal and regulatory frameworks requiring reform.
Running 4 PPIE workshops to understand their priorities and beliefs around safeguarding and securing data
Developing a set of recommendations including
suggested open-source toolsets for TREs to use to measure and reduce disclosure risk
descriptions of the technical and legal controls and policies TREs should implement across the 5 Safes to support AI algorithm disclosure control
training implications for both TRE staff and how they validate researchers
Conclusion
GRAIMatter is developing a set of usable recommendations for TREs to use to guard against the additional risks when disclosing trained AI models from TREs.
| Original language | English |
|---|---|
| Title of host publication | Conference Proceedings for International Population Data Linkage Conference 2022 |
| Volume | 7 |
| Edition | 3 |
| DOIs | |
| Publication status | Published - 25 Aug 2022 |
| Event | International Population Data Linkage Conference 2022 - Edinburgh Duration: 7 Sept 2022 → 9 Sept 2022 Conference number: 2022 https://ipdln.org/2022-conference |
Conference
| Conference | International Population Data Linkage Conference 2022 |
|---|---|
| Abbreviated title | IPDLN |
| City | Edinburgh |
| Period | 7/09/22 → 9/09/22 |
| Internet address |
Fingerprint
Dive into the research topics of 'GRAIMatter: Guidelines and Resources for AI Model Access from TrusTEd Research environments'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Green Paper: Disclosure Control Of Al Algorithms From TREs (GRAIMATTER) lead: University of Dundee other instns: National Services Scotland, University of the West England UWE - Bristol, University of Edinburgh, URV (Spain)
Daly, A. (Investigator), Jefferson, E. (Investigator) & Malone, M. (Investigator)
4/01/22 → 31/08/22
Project: Research
Research output
- 1 Other report
-
GRAIMATTER Green Paper: Recommendations for disclosure control of trained Machine Learning (ML) models from Trusted Research Environments (TREs)
Jefferson, E. (Lead / Corresponding author), Reel, S. (Lead / Corresponding author), Crespi Boixader, A., Roche, M., Mansouri-Benssassi, E., Beggs, J., Chuter, A., Cole, C., Ritchie, F., Daly, A., Smith, J., Liley, J., Kerasidou, K., Tava, F., McCarthy, A., Preen, R., Blanco-Justicia, A., Mansouri-Benssassi, E., Domingo-Ferrer, J. & Rogers, S. & 1 others, , 21 Sept 2022, Zenodo: DARE UK. 99 p.Research output: Book/Report › Other report
Open AccessFile247 Downloads (Pure)