DARE UK (Data and Analytics Research Environments UK)
Published September 21, 2022 | Version v1
Report Open

GRAIMATTER Public Summary: Recommendations for disclosure control of trained Machine Learning (ML) models from Trusted Research Environments (TREs)

Creators

  • 1. Bristol Business School, University of the West of England, Bristol
  • 2. Public/patient advocate, University of Dundee, Dundee.
  • 3. Division of Population Health and Genomics, School of Medicine, University of Dundee, Dundee.
  • 4. Department of Health and Social SciencesUniversity of the West of England, Bristol
  • 5. Department of Computer Science and Creative Technologies, University of the West of England, Bristol
  • 6. Department of Computer Engineering and Mathematics, Universitat Rovira i Virgili, Tarragona, Catalonia
  • 7. Division of Population Health and Genomics, School of Medicine, University of Dundee.
  • 8. Department of Mathematical Sciences, Durham University.
  • 9. Dundee Law School School of Humanities Social Sciences and Law, University of Dundee.
  • 10. Leverhulme Research Centre for Forensic Science, School of Science and Engineering, University of Dundee.
  • 11. NHS National Services Scotland.
  • 12. Division of Population Health and Genomics, School of Medicine, University of Dundee. & University of Glasgow

Description

GRAIMATTER has developed a draft set of usable recommendations for TREs to guard against the additional risks when disclosing trained AI models from TREs. This report provides a summary of our recommendations for a general public audience. The detailed Green Paper on recommendations can be found at  DOI: 10.5281/zenodo.7089491

If you would like to provide feedback or would like to learn more, please contact Smarti Reel (sreel@dundee.ac.uk) and Emily Jefferson (erjefferson@dundee.ac.uk).

 

 

 

Files

GRAIMATTER Public Summary.pdf

Files (424.2 kB)

Name Size Download all
md5:023f76c873ea4f971af9277deec3b842
424.2 kB Preview Download

Additional details

Funding

Guidelines and Resources for AI Model Access from TrusTEd Research environments (GRAIMatter) MC_PC_21033
UK Research and Innovation