Jill-Jênn Vie

ML Researcher in RIKEN AIP, Tokyo

Hey! Come to Montréal on June 12 to our Optimizing Human Learning workshop about adaptive sequences for learning.
Register now! See the program! Share on Twitter!

I’m a postdoctoral researcher at RIKEN AIP in the Human Computation team working with Professor Hisashi Kashima.
I also write free software for the French government (Vie, Popineau, Tort, et al. 2017).

My résumé: English / French
GitHub projects: jilljenn
vie@jill-jenn.net (check my YouTube channel if you like piano)

Research Interests

Interactive Machine Learning

Adaptive algorithms, optimal design (D-optimality), interactive recommender systems (like anime? try Mangaki)

Feature Extraction, Representation Learning

Using posters for manga and anime recommendations (Vie et al. 2017), deep factorization machines (Vie 2018), variational autoencoders

Learning Analytics, Knowledge Tracing

How to use the logs of educational platforms (MOOCs, PIX) to optimize human learning?
Multidimensional item response theory, cognitive diagnosis [slides], multistage testing [slides] [code]

We are presenting Knowledge Tracing Machines [poster] at AIP-IIS-MLGT 2018 in Atlanta.
We are presenting Knowledge Tracing Machines [poster] at AIP-IIS-MLGT 2018 in Atlanta.
Our paper Using Posters to Recommend Anime and Mangas in a Cold-Start Scenario [arXiv] [slides] has been accepted to MANPU 2017 in Kyoto.
Our paper Using Posters to Recommend Anime and Mangas in a Cold-Start Scenario [arXiv] [slides] has been accepted to MANPU 2017 in Kyoto.
Our article Automated Test Assembly using DPPs for Handling Learner Cold-Start in Large-Scale Assessments has been accepted in the journal IJAIED 2018.
Our article Automated Test Assembly using DPPs for Handling Learner Cold-Start in Large-Scale Assessments has been accepted in the journal IJAIED 2018.

Achievements

Teaching algorithms

2 books about algorithms & data structures (English version soon)
1 Python package: pip install tryalgo128 algorithms for coding contests
✅ A data challenge with Kyoto University. [problem] [solutions]

… to everyone

✅ A programming summer school for K-12 girlsGirls Can Code! since 2014
✅ A TV show about algorithms that take control of our lives → Blame the Algorithm
✅ An open-source project: anime recommender systemMangaki

Publications

See my Scholar page

Books

I co-wrote two books about algorithms and data structures: preparing programming contests and coding interviews (Dürr and Vie 2016) and (Belghiti, Mansuy, and Vie 2016).
The book is in French but the code is in Python: all 128 algorithms are on GitHub, in the Python package tryalgo.

Book Chapters

I wrote a chapter about adaptive assessment (Vie, Popineau, Bourda, et al. 2017) in a learning analytics book, where I identify similarities between cognitive diagnostic models and item response theory.

Conference Proceedings

I used to do cryptography in my master’s degree and published an article about leakage resilience (Abdalla and Vie 2012).
Used a General Diagnostic Model for adaptive testing and showed that it outperformed the DINA and Rasch models at predicting examinee performance (Vie et al. 2016).


Abdalla, Michel, and Jill-Jênn Vie. 2012. “Leakage-Resilient Spatial Encryption.” In International Conference on Cryptology and Information Security in Latin America, 78–99. Springer.

Belghiti, Ismael, Roger Mansuy, and Jill-Jênn Vie. 2016. Les clés pour l’info : ENS et Agrégation (option D). Calvage et Mounet.

Dürr, Christoph, and Jill-Jênn Vie. 2016. Programmation efficace. Ellipses.

Vie, Jill-Jênn. 2018. “Deep Factorization Machines for Knowledge Tracing.” In The 13th Workshop on Innovative Use of NLP for Building Educational Applications. https://arxiv.org/abs/1805.00356.

Vie, Jill-Jênn, and Hisashi Kashima. 2018. “Knowledge Tracing Machines.” Presented at the AIP-IIS-MLGT workshop at Georgia Tech, Atlanta, GA on March 8, 2018. https://github.com/jilljenn/ktm.

Vie, Jill-Jênn, Fabrice Popineau, Yolaine Bourda, and Éric Bruillard. 2016. “Adaptive Testing Using a General Diagnostic Model.” In European Conference on Technology Enhanced Learning, 331–39. Springer.

———. 2017. “A Review of Recent Advances in Adaptive Assessment.” In Learning Analytics: Fundaments, Applications, and Trends, 113–42. Springer.

Vie, Jill-Jênn, Fabrice Popineau, Éric Bruillard, and Yolaine Bourda. 2018a. “Automated Test Assembly for Handling Learner Cold-Start in Large-Scale Assessments.” International Journal of Artificial Intelligence in Education. Springer, 1–16. https://rdcu.be/G30H.

———. 2018b. “Utilisation de tests adaptatifs dans les MOOC dans un cadre de crowdsourcing.” Revue STICEF, Volume 24, numéro 2, 2017. https://doi.org/10.23709/sticef.24.2.6.

Vie, Jill-Jênn, Fabrice Popineau, Jean-Bastien Grill, Éric Bruillard, and Yolaine Bourda. 2015a. “Predicting Performance over Dichotomous Questions: Comparing Models for Large-Scale Adaptive Testing.” In 8th International Conference on Educational Data Mining.

———. 2015b. “Prédiction de performance sur des questions dichotomiques : comparaison de modèles pour des tests adaptatifs à grande échelle.” In Atelier Évaluation des Apprentissages et Environnements Informatiques.

Vie, Jill-Jênn, Fabrice Popineau, Françoise Tort, Benjamin Marteau, and Nathalie Denos. 2017. “A Heuristic Method for Large-Scale Cognitive-Diagnostic Computerized Adaptive Testing.” In Proceedings of the Fourth (2017) Acm Conference on Learning @ Scale, 323–26. ACM. https://github.com/jilljenn/las2017-wip/blob/master/poster-las2017.pdf.

Vie, Jill-Jênn, Florian Yger, Ryan Lahfa, Basile Clement, Kévin Cocchi, Thomas Chalumeau, and Hisashi Kashima. 2017. “Using Posters to Recommend Anime and Mangas in a Cold-Start Scenario.” In 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), 03:21–26. https://arxiv.org/abs/1709.01584.