News
December 2023
- 20/12: Our paper "Probabilistic Dataset Reconstruction from Interpretable Models" has been accepted to the 2nd IEEE Conference on Secure and Trustworthy Machine Learning (SATML 2024). Preprint is available on arXiv and HAL!
November 2023
- 20/11: The final version of my PhD thesis is now ready! It is available here.
October 2023
- 09/10: I defended my PhD thesis, and I am glad to announce that I am now a Doctor of Philosophy in compter science. The recording of my presentation is available online.
July 2023
- 03/07 - 07/07: I attended the French Artificial Intelligence Platform (PFIA 2023) in Strasbourg (France). I presented our accepted full paper on leveraging a model's fairness to enhance the reconstruction of its training set sensitive attributes within the Meeting of Young Researchers in Artificial Intelligence.
February 2023
- 20/02 - 24/02: I attended the 24th annual congress of the société Française de Recherche Opérationnelle et d'Aide à la Décision, from the 20th to the 24th February in Rennes (France). Detailed program is available here. I presented our accepted extended abstract within the Operational Research and Machine Learning track. Slides for my presentation are available here.
- 08/02 - 10/02: I attended the very first IEEE Conference on Secure and Trustworthy Machine Learning (SATML) in Raleigh (North Carolina, USA). I presented our work on exploiting the fairness of a trained model to enhance an adversary's ability to reconstruct the model's training set sensitive attributes (recording available here). The full article is available online on the conference website or on ArXiV. I also participated in a poster session, which included all accepted papers. My poster can be found here.
- 07/02 - 14/02: Mohamed Siala attended the 37th AAAI conference on Artificial Intelligence to present our Machine Learning journal article within the AAAI Journal Track as well as the Bridge on Constraint Programming and Machine Learning. The recording of the presentation can be found here, the slides here and the poster here.
January 2023
- We have open internship positions! More details here. Please e-mail me to apply or if you need more information!
December 2022
- Our Machine Learning journal paper "Improving Fairness Generalization Through a Sample-Robust Optimization Method" has been accepted to the AAAI-23 Constraint Programming and Machine Learning Bridge.
November 2022
- Our Machine Learning journal paper "Improving Fairness Generalization Through a Sample-Robust Optimization Method" has been accepted to the AAAI-23 journal track. In addition to the poster presentation, it has also been selected for oral presentation.
- Our paper "Exploiting Fairness to Enhance Sensitive Attributes Reconstruction" has been accepted to the 1st IEEE Conference on Secure and Trustworthy Machine Learning (SATML 2023). Preprint is available on arXiv and HAL! Check out the list of accepted papers on the conference's website.
- I participated in the organization of the LAAS-CNRS Doctoral Students day
September 2022
- Our full paper "Learning Optimal Fair Scoring Systems for Multi-Class Classification" has been accepted for publication in the 34th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2022). Preprint is available on HAL!
July 2022
- The recording of my presentation at the CPAIOR 2022 conference is now available on YouTube.
- 07/07: I gave a local seminar for the DS4DM Coffee Talks at Polytechnique Montréal. I presented an overview of my past and current research topics on the intersections between fairness, interpretability, and privacy in Machine Learning. I then detailed my work on leveraging ILP to learn optimal fair rule lists. Recording of my talk is available on YouTube.
- 06/07: Our article "Improving Fairness Generalization Through a Sample-Robust Optimization Method" is published in the Special Issue on Safe and Fair Machine Learning within the Machine Learning journal. The paper is available on the journal's website. A view-only version can also be accessed for free through Springer Nature SharedIt. Finally, the preprint can be found on HAL.
June 2022
- 20/06 - 23/06: I attended the 19th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research (CPAIOR 2022), in Los Angeles (California, USA). I presented our accepted full paper "Leveraging Integer Linear Programming to Learn Optimal Fair Rule Lists". Our article is available in the conference's proceedings (or on HAL).
May 2022
- 25/05: I gave a local seminar at MILA (Montreal). I presented an overview of my past and current research topics on the intersections between fairness, interpretability, and privacy in Machine Learning. I then detailed my work on leveraging ILP to learn optimal fair rule lists.
- I attended Calcul Québec's 2022 Spring school tutorial on GPU Programming with Python. We experienced just-in-time CUDA compilation with the Numba Python library.
- Our article "Improving Fairness Generalization Through a Sample-Robust Optimization Method" has been accepted for publication in the Special Issue on Safe and Fair Machine Learning within the Machine Learning journal. Preprint is available on HAL.
- 11/05: I gave a LATECE (Laboratoire de Recherches Transdisciplinaires sur les Ecosystèmes Informatiques) seminar at UQAM (Montreal). I presented an overview of my past and current research topics on the intersections between fairness, interpretability, and privacy in Machine Learning. I then detailed my work on leveraging ILP to learn optimal fair rule lists. Slides for my presentation are available here.
- 09/05 - 10/05: I attended the PrivSec team seminar at UQAM, Montreal (Canada). I presented my work on the use of ILP to enhance interpretable fair learning.
- 05/05 - 07/05: I attended the TimeWorld Global Congress on Artificial Intelligence in Montreal (Canada).
- 02/05: I just arrived in Montreal (Canada), where I will be visiting Prof. Sébastien Gambs at UQAM and Prof. Ulrich Aïvodji at ETS, until the end of July. I will work on different topics intersecting with privacy in machine learning, and participate in several local events.
February 2022
- Our full paper "Leveraging Integer Linear Programming to Learn Optimal Fair Rule Lists" has been accepted for publication at the 19th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research (CPAIOR 2022). Preprint is available here, and the list of accepted papers is provided on the conference's website.
- We released FairCORELSV2, an enhanced version of the FairCORELS algorithm embedding the contributions that we will present at the CPAIOR 2022 conference. In a nutshell, FairCORELSV2 proposes a novel pruning method, leveraging Mixed Integer Linear Programming to prune the search space of fair rule lists efficiently, considering jointly the objective function and the fairness constraint. It also provides a new prefix permutation map that can be used to speed up the learning process while preserving the guarantee of optimality.
- I attended the 23rd annual congress of the société Française de Recherche Opérationnelle et d'Aide à la Décision, from the 23rd to the 25th February in Lyon (France). Detailed program is available here. I presented our accepted extended abstract within the Operational Research and Machine Learning track. Slides for my presentation are available here.
January 2022
- I joined the NSERC (Natural Sciences and Engineering Research Council of Canada) CREATE on Responsible AI program. This multidisciplinary training program is developped by a consortium of researchers from Ryerson, McGill, Western, Waterloo and UQAM. It covers three main areas: AI Ethics by Design, Privacy-enhanced Analytics, and AI Accountability.
November 2021
- 01/11 - 05/11: I attended the30th ACM International Conference on Information and Knowledge Management (CIKM 2021). The conference was hosted in Gold Coast (Queensland, Australia) and took place online. I presented our accepted paper during two poster sessions.CIKM 2021 proceedings are online in the ACM Digital Library. Our article, along with a recorded video presentation of our work, is available here.
October 2021
- 05/10 - 07/10: I attended the DO department workshop in Mauvezin. Members of the three teams of our department (DISCO, MAC and ROC) presented their topics of research. I presented my work on statistical fairness generalization.
September 2021
- 27/09 - 01/10: I attended the GDR IA Autumn shool on Explainable IA (IA²), at the Sorbonne Center for Artificial Intelligence (Université Pierre et Marie Curie), in Paris. I presented my PhD topics during two poster sessions.
- I have been granted a SIGIR Student Travel Award to attend the30th ACM International Conference on Information and Knowledge Management (CIKM 2021).
August 2021
- Our demo paper "FairCORELS, an Open-Source Library for Learning Fair Rule Lists" was accepted for presentation and inclusion in the proceedings of the30th ACM International Conference on Information and Knowledge Management (CIKM 2021).
July 2021
I attended the 19th Meeting of Young Researchers in Artificial Intelligence (within the French Artificial Intelligence Platform), from the 1st to the 2nd July. Detailed program is available here.
I presented our accepted paper (proceedings are available here) within the Fundamental AI session.
May 2021
I attended the following Calcul Québec's 2021 Spring school workshops:
- Parallel Programming with MPI
- Parallel Programming with OpenMP
Knowledge of these standards facilitate the exploitation of massively parallel computing infrastructures, such as those offered by Compute Canada.
April 2021
I attended the 22nd annual congress of the société Française de Recherche Opérationnelle et d'Aide à la Décision, from the 26th to the 30th April. Detailed program is available here.
I presented our accepted extended abstract within the Operational Research and Machine Learning track. Slides for my presentation are available here.
November 2020
I attended the Joint ACP, ANITI, CNRS GDR IA, GDR RO International Autumn school on Combinatorial Optimization, Constraint Programming and Machine Learning, from the 23rd to the 27th November. Detailed program is available here.
I also participated in the organization of the Autumn school's Constraint Programming Hackathon consisting in modeling and breaking the well-known Enigma machine.