Gianni Franchi

I am an Assistant Professor in the Department of Computer Science and Robotics at ENSTA Paris, part of Institut Polytechnique de Paris, France. I also represent the AI Jean Zay user council. Feel free to connect with me for any inquiries.

I have been teaching Deep Learning, Computer Vision, and Machine Learning courses at both ENSTA Paris and Télécom Paris since 2020. Before my current role, I held positions as a postdoctoral researcher at Paris Saclay University from 2018 to 2020, and previously at Seigen University from 2016 to 2018. I earned my PhD in 2016 from Mines de Paris under the mentorship of Jesus Angulo, focusing on Fusion of Information, Machine Learning, and Image Processing.

My research interests concern machine learning and computer vision applied to all sorts of modalities. I focus particularly on Uncertainty Quantification and XAI.

Email  /  CV  /  Scholar  /  Twitter  /  Github

profile photo

Current PhD Students

  • 2022 – 2025, Rémi Kazmierczak, co-advised with Eloïse Berthier, Goran Frehse, topic: XAI and foundation models
  • 2022 – 2025, Olivier Laurent, co-advised with Adrien Chan Hon Tong, Emanuel Aldea, topic: Uncertainty and Deep Learning
  • 2022 – 2025, Adrien Lafage, co-advised with Mathieu Barbier, David FILLIAT, topic: Uncertainty and trajectory forecasting
  • 2022 – 2025, Mouïn Ben Ammar, co-advised with Nacim Belkhir, Antoine Manzanera, topic: Anomaly detection

Alumni Students

  • 2020 – 2023, Xuanlong Yu, co-advised with Emanuel Aldea

Postdocs

  • 2021 – 2024, Antoine Guillaume

Research

I'm interested in robust in robust computer vision, anomaly detection, uncertainty quantification, out-of-distribution detection, certifiable AI, and explainable AI. If you have any projects or questions where collaboration could be beneficial, please don't hesitate to reach out to me.

Additionally, I'm currently involved with a group in the development of a PyTorch library Torch Uncertainty tailored for uncertainty quantification. If you're interested in contributing to this project, your participation would be greatly appreciated.


Feel free to steal this website's source code. Do not scrape the HTML from this page itself, as it includes analytics tags that you do not want on your own website — use the github code instead. Also, consider using Leonid Keselman's Jekyll fork of this page.