2026
BA-FedSHAP: A reproducible toolkit for auditing background-induced attribution drift
SSRN
Preprint on how the choice of background distribution alters federated attribution results, and what that does to an audit.
PublicationsAuthor
Policy
Federated learning for children's data: encoding trust in education systems
- Year
- 19 May 2025
- Role
- Author
- Venue
- UNICEF Innocenti, Office of Strategy and Evidence
Argues that governments can use sensitive education data without centralising it, and what governance that requires.
Authors: Roy Saurabh
Evidence
Sources
What each document establishes, and where to read it.
UNICEF Innocenti
Federated learning for children's data: encoding trust in education systems
Article authored by Roy Saurabh and published by UNICEF Innocenti on 19 May 2025, listed among the outputs of the good-governance-of-children's-data project.
UNICEF Innocenti
Good governance of children's data, advisory groups
Names Roy Saurabh among the current expert advisory group members shaping the second phase of UNICEF Innocenti's children's data governance project.
Related work
Related work
2026
BA-FedSHAP, auditing background-induced attribution drift
Measures how much a federated explanation depends on an arbitrary modelling choice.
Research softwareAuthor
2023
The Transformative Potential of Data for Learning
Broadband Commission for Sustainable Development
Broadband Commission working-group report on the governance of education data, written by the UNESCO team supporting the group.
ReportsWriting team
2022
The Transformative Potential of Data for Learning, interim report
Broadband Commission for Sustainable Development
Preliminary findings of the working group. Roy was on the UNESCO writing team for this edition too.
ReportsWriting team