EVENT

JFLI seminar by Prof. Éric Monfroy & Prof. Arthur Charpentier

We are pleased to inform you about the upcoming seminar by Éric Monfroy, professor at University of Angers, France and Arthur Charpentier, Professor at Université du Québec, Montréal, Canada Everyone interested is cordially invited to attend!

Title:

From grammar inference to Good anomaly detection

Speaker:

Éric Monfroy, professor at University of Angers, France

Abstract:

Learning grammar, or grammatical inference, consists in generating a grammar (generally as a NFA - Non-deterministic Finite Automaton) from some positive words (to be recognized by the NFA) and negative words (to be rejected by the NFA). I will present some work to generate NFA with constraint programming (i.e., SAT) and to improve the models and the automata. Next, the current work on this topic will give some ideas for enhancing the "quality" of the recognition process. I will end with some future work, and how to "good" anomaly detection.

Time:

14:00-15:00

Title:

Counterfactual and Transport-Based Methods for Understanding Indirect Discrimination in Algorithmic Systems

Speaker:

Arthur Charpentier, Professor at Université du Québec, Montréal, Canada

Abstract:

Understanding disparities between demographic groups in algorithmic predictions remains a central challenge in responsible AI. Classical decomposition methods such as the Kitagawa-Oaxaca-Blinder framework, recently extended to nonlinear and machine-learning settings by Tierney et al. (AAAI 2026), show that observed gaps may arise either from legitimate differences in feature distributions or from structural bias. However, such aggregate decompositions provide limited insight into individual-level counterfactual behaviour. In this talk, I will present recent methodological advances that combine causal reasoning with optimal transport to characterize direct and indirect discriminatory pathways in modern predictive systems. Building on transport-based counterfactuals (Fernandes Machado et al., AAAI 2025; IJCAI 2025), we obtain individual-level counterfactual mediators that respect a given causal graph, including both continuous and categorical variables. This enables a fine-grained decomposition of model disparities into components attributable to causal pathways--beyond what is possible with standard fairness metrics or feature-removal strategies. The presentation will emphasize: the connection between decomposition-based fairness analyses and causal mediation; the construction of transport-based counterfactuals aligned with probabilistic graphical models; and applications showing how indirect discrimination can propagate through proxy variables even when sensitive features are not used. The goal is to give a concise and technically grounded overview of how optimal transport and counterfactual inference can provide interpretable tools for understanding fairness issues in machine-learning models. This talk is intended for researchers interested in causal ML, fairness analysis, and transport-based generative methods.

Time:

15:00-16:00

Time/Date:

14:00-16:00 /Wednesday,December 3 , 2025

Place:

Room 1810, NII

Contact:

If you would like to join, please contact by email.
Email : megkaneko[at]nii.ac.jp

Link:

Japanese-French Laboratory for Informatics