James Liley

Improving the quality of machine learning in health applications and clinical research

2020 · 65 Zit. · Nature Machine Intelligence

Model updating after interventions paradoxically introduces bias

2020 · 7 Zit. · arXiv (Cornell University)

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

2022 · 5 Zit. · arXiv (Cornell University)

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

2022 · 2 Zit. · Zenodo (CERN European Organization for Nuclear Research)

Ethical considerations of use of hold-out sets in clinical prediction model management

2024 · 2 Zit. · AI and Ethics

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

2022 · 0 Zit. · Discovery Research Portal (University of Dundee)

Ethical considerations of use of hold-out sets in clinical prediction model management

2024 · 0 Zit. · arXiv (Cornell University)

Safe machine learning model release from Trusted Research Environments: The SACRO-ML package

2022 · 0 Zit. · arXiv (Cornell University)

Holdout sets for safe predictive model updating

2025 · 0 Zit. · The Annals of Applied Statistics

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

2022 · 0 Zit. · Zenodo (CERN European Organization for Nuclear Research)