About me

I hold a PhD in data science from Université Paris-Saclay & CEA-List. During my PhD, I focused on the effectiveness and the security of federated learning models while dealing with heterogeneous data. I was advised by Cédric Gouy-Pailler and Aurélien Mayoue.

On the job market!

I defended my PhD thesis in February 2024. I decided to take a break from research (and work in general) and since I have always loved traveling, I visited different countries in Asia for more than 3 months. You can check my trip here :)

After that break, I am now looking for a job as a data scientist/research scientist. I would like to use AI to solve real-world problems and have a positive impact on our society.

News!

• February 2024: Defended my PhD thesis!
• January 2024: G-Research grant awarded to participate at ICASSP 2024!
• December 2023: Conference paper accepted at ICASSP 2024!
• November 2023: Received the best paper award at IEEE TPS - ISA, in Atlanta!
• September 2023: Conference paper accepted at IEEE TPS - ISA.
• June 2023: Conference paper accepted at local French conference GRETSI.
• March 2023: Short talk at the GT (groupe de travail) on cybersecurity, Hub France IA.
• January 2023: Poster presented at the Workshop FL-Day - Decentralized Federated Learning, DATAIA Institute of Paris-Saclay.
• June 2022: Short talk at the GDR (groupe de recherche - research group) on distributed and federated AI, Sorbonne Université.
• May 2022: Poster presented at the ICST (Information and Communication Sciences and Technologies) doctoral students’ day.
• April 2022: (first) Accepted conference paper at IJCNN 2022!
• November 2023: Started my PhD!
• September 2020: Finished my internship at EDF R&D, concluding my studies at Telecom SudParis! My internship focused on the use of deep learning to forecast wind power production.

Publications

• Espinoza Castellon, F., Montesuma, E. F., Mboula, F. N., Mayoue, A., Souloumiac, A., & Gouy-Pailler, C. (2024, April). Federated dataset dictionary learning for multi-source domain adaptation. In 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 5610-5614). IEEE.
• Espinoza Castellon, F., Singh, D., Mayoue, A., & Gouy-Pailler, C. (2023, November). FUBA: Federated Uncovering of Backdoor Attacks for Heterogeneous Data. In 2023 5th IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA) (pp. 55-63). IEEE.
• Espinoza Castellon, F., Singh, D., Mayoue, A., & Gouy-Pailler, C. (2023, August). Défense contre les attaques par porte dérobée en apprentissage fédéré par estimation du motif d’attaque et élagage. In 29ème Colloque sur le traitement du signal et des images (p. 653-656). GRETSI - Groupe de Recherche en Traitement du Signal et des Images.
• Espinoza Castellon, F., Mayoue, A., Sublemontier, J. H., & Gouy-Pailler, C. (2022, July). Federated learning with incremental clustering for heterogeneous data. In 2022 International Joint Conference on Neural Networks (IJCNN) (pp. 1-8). IEEE.

Teaching

Travaux Pratiques sur les attaque adversariales ici.