Comparative evaluation procedure for n. 1 post-doc research contract - Department of Computer Science and Engineering
Scientific-Disciplinary Group
09/IINF-05 - Information Processing Systems
Description
Title of the project: "Transfer and Federated Learning for HPC Monitoring and Optimization". The research will develop advanced transfer and federated learning methods for predictive maintenance, and energy-aware optimization in High-Performance Computing environments. Key objectives include: (i) designing transfer learning strategies robust to hardware and workload heterogeneity across HPC systems (domain adaptation, selective fine-tuning); (ii) developing communication-efficient federated protocols for non-IID telemetry data, with adaptive aggregation and client selection; (iii) integrating these techniques into an emerging HPC monitoring platform, including data ingestion pipelines, distributed training orchestration and lifecycle support; and (iv) defining validation benchmarks and operational KPIs to support adoption across HPC centers.
Compensation
28,456 Euro
Job posting website
Number of positions
1
Maximum duration
36.0
Funding body
Università di Bologna
How to apply
Other
Selection process
Click to expand
View the original posting on the MUR website: Go to MUR website