Scholar
Eros Fanì
Google Scholar ID: rwto7AgAAAAJ
Basque Center for Applied Mathematics
Citations & Impact
Citations
163
H-index
5
i10-index
2
Publications
12
Co-authors
18
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published and presented work at several international peer-reviewed conferences and workshops including ICML, NeurIPS, CVPR, IROS, WACV, and to an IEEE journal; Honored as an Outstanding Reviewer at ECCV in 2024; Paper “Diversity-Driven Learning: Tackling Spurious Correlations and Data Heterogeneity in Federated Models” accepted at FLTA25; Paper “Interaction-Aware Gaussian Weighting for Clustered Federated Learning” accepted at ICML25; Paper “Resource-Efficient Personalization in Federated Learning with Closed-Form Classifiers” accepted at IEEE Access.
Research Experience
- Machine Learning Researcher at Gensyn, focusing on massively decentralized and heterogeneous distributed learning (Oct 2025 - present); Postdoctoral Fellow & Software Engineer at Basque Center for Applied Mathematics, Lead developer of Act.AI project (Nov 2024 - Sep 2025); MSc thesis supervisor at Politecnico di Torino, Topic: Clustered Federated Learning (Sep 2023 - Oct 2024); Teaching Assistant at Politecnico di Torino, teaching multiple courses (A.Y. 2022/23); MSc thesis supervisor at Politecnico di Torino, Topic: Federated Visual Geo-Localization (Sep 2022 - Jul 2023); Teaching collaborator as an undergrad at Politecnico di Torino (A.Y. 2018/19).
Education
- Ph.D. in Computer and Control Engineering, Politecnico di Torino, Advisor: Prof. B. Caputo; Thesis Title: Addressing Heterogeneity in Federated Learning for Real-world Vision Applications. The Ph.D. project was conducted within the European Laboratory for Learning and Intelligent Systems (ELLIS) Ph.D. program.
Background
- Research Interests: Building scalable, distributed models over uniquely decentralized and heterogeneous infrastructure, and addressing the challenge of data heterogeneity in real-world applications of federated learning. Areas of expertise include semantic segmentation, domain adaptation, domain generalization, and semi-supervised learning.
Miscellany
- Personal interests not provided.
