Jill-Jênn Vie

Researcher at Inria

I am a research scientist at Inria interested in online factorization, deep generative models and educational applications of machine learning. I have taught deep learning and algorithms, complexity and programming.

Feel free to contact me by mail or Twitter.

Research Interests

Deep generative models of time series

If we can generate assessment data from MOOCs, we can predict, explain and optimize human learning.
If we can do this with privacy guarantees, we can share this dataset with more users.
See our tutorial about knowledge tracing and our workshop WASL 2020 @ AIED [video].

Recommender systems with side information

How to model uncertainty and side information in preference elicitation? See our demo Mangaki and keynote.

Our article Knowledge Tracing Machines has been presented at AAAI 2019. See also our code & tutorial (Vie and Kashima 2019).
Our paper Using Posters to Recommend Anime and Mangas in a Cold-Start Scenario [slides] has been accepted to MANPU 2017.
We received the Best Paper Award at EDM 2019 for our learning/forgetting student model DAS3H (Choffin et al. 2019).

Other Achievements

✅ I designed and implemented free software that is used to certify the digital skills of every French citizen (Pix)
✅ We founded a programming summer school for K-12 girlsGirls Can Code! running since 2014
✅ See my projects and CV


Selected Publications

See all publications / My Scholar page

Bergner, Yoav, Peter Halpin, and Jill-Jênn Vie. 2021. “Multidimensional Item Response Theory in the Style of Collaborative Filtering.” Psychometrika, in press.
Choffin, Benoı̂t, Fabrice Popineau, Yolaine Bourda, and Jill-Jênn Vie. 2019. DAS3H: Modeling Student Learning and Forgetting for Optimally Scheduling Distributed Practice of Skills.” In Proceedings of the Twelfth International Conference on Educational Data Mining (EDM 2019), 29–38. https://arxiv.org/abs/1905.06873.
Vie, Jill-Jênn, and Hisashi Kashima. 2019. Knowledge Tracing Machines: Factorization Machines for Knowledge Tracing.” In Proceedings of the 33th AAAI Conference on Artificial Intelligence, 750–57. https://arxiv.org/abs/1811.03388.