2026_IR_16_DEI – Federated Learning Methods for Biomedical Data: Fairness
Deadline: 16/07/2026
Scientific-Disciplinary Group
09/IINF-05 - Information Processing Systems
Description
The researcher will investigate fairness-aware federated learning approaches for multimodal biomedical data within the HEREDITARY project (Horizon Europe RIA, grant 101137074), with focus on the Multimodal Analytics and Learning Platform (WP4). The activity addresses statistical bias and representational unfairness that arise when training predictive models over federated clinical data with heterogeneous demographic composition across sites. The main tasks include: (i) formalizing and operationalizing fairness criteria — demographic parity, equalized odds, calibration — in the federated non-IID setting where individual-level data cannot be shared; (ii) designing bias detection and mitigation algorithms, such as fairness-constrained federated optimization and post-processing debiasing, applicable to tabular clinical data, genomic data, and medical imaging...
Job posting website
Funding body
Dipartimento di Ingegneria dell'Informazione - Università degli studi di Padova
How to apply
View the original posting on the MUR website: Go to MUR website