All posts by Arthur Charpentier

Arthur Charpentier, professor in Montréal, in Actuarial Science. Former professor-assistant at ENSAE Paristech, associate professor at Ecole Polytechnique and assistant professor in Economics at Université de Rennes 1.  Graduated from ENSAE, Master in Mathematical Economics (Paris Dauphine), PhD in Mathematics (KU Leuven), and Fellow of the French Institute of Actuaries.

Workshop on fairness and discrimination in insurance (registration is open)

Almost two years ago, on May 13th 2022, we organized a Workshop on fairness and discrimination in insurance, JEDA’22, at Laval University (in Québec city), with Marie-Pier Côté.

It was a beautiful sucess, with a lot of persons in person, for one of the first event post-pandemic. The second workshop (JEDA’24) will be organized in less than a month, on May 16th.

Registrations are open ! We will have in the room Fei Huang (UNSW Sidney), David Schraub (Chicago Actuarial Association), Marie-Ève Lainez, Autorité des marchés financiers), Laurence Barry (Chaire PARI), Agathe Fernandes Machado (UQÀM) Mallika Bender (Casualty Actuarial Society), Christopher Cooney (TD Insurance) and Olivier Côté (Université Laval).

Groupe de travail, dimension territoriale dans la mutualisation des risques climatiques

Demain matin, on m’a demandé d’intervenir au groupe de travail “mutualisation des risques” de France Stratégie. A priori, j’interviendrais dans la discussion sur “Comment mutualiser les risques climatiques entre territoires ? “, et j’ai préparé quelques slides sur la base de mes travaux depuis une petite vingtaine d’année (ma participation au groupe de travail du Commissariat au Plan sur L’État et l’assurance des risques nouveaux remonte à 2005…).

Geospatial Disparities: A Case Study on Real Estate Prices in Paris

Our paper, Geospatial Disparities: A Case Study on Real Estate Prices in Paris, and Agathe Fernandes Machado, François Hu, Philipp Ratz and Ewen Gallic, is now online on ArXiv,

Driven by an increasing prevalence of trackers, ever more IoT sensors, and the declining cost of computing power, geospatial information has come to play a pivotal role in contemporary predictive models. While enhancing prognostic performance, geospatial data also has the potential to perpetuate many historical socio-economic patterns, raising concerns about a resurgence of biases and exclusionary practices, with their disproportionate impacts on society. Addressing this, our paper emphasizes the crucial need to identify and rectify such biases and calibration errors in predictive models, particularly as algorithms become more intricate and less interpretable. The increasing granularity of geospatial information further introduces ethical concerns, as choosing different geographical scales may exacerbate disparities akin to redlining and exclusionary zoning. To address these issues, we propose a toolkit for identifying and mitigating biases arising from geospatial data. Extending classical fairness definitions, we incorporate an ordinal regression case with spatial attributes, deviating from the binary classification focus. This extension allows us to gauge disparities stemming from data aggregation levels and advocates for a less interfering correction approach. Illustrating our methodology using a Parisian real estate dataset, we showcase practical applications and scrutinize the implications of choosing geographical aggregation levels for fairness and calibration measures.

Insurance Data Science, IDSC’24 (in Stockholm)

This year, the Insurance Data Science Conference will take place in Stockholm. The abstract booklet with all accepted speakers at IDSC 2024 is now available. Agathe will present some recent joint work on calibration (inspired by arXiv:2402.07790 – as well as more recent work)

I will also attend, to present some work on fairness, with multiple sensitive attributes (based on arXiv.2309.06627 and another article that should be finalized by then about our equipy python package),

Boarding for ISS: Imbalanced Self-Supervised Discovery of a Scaled Autoencoder for Mixed Tabular Datasets

Our paper “Boarding for ISS: Imbalanced Self-Supervised Discovery of a Scaled Autoencoder for Mixed Tabular Datasets,” by Samuel Stocksieker, co-autored with Denys Pommeret, has been accepted for presentation at the IEEE World Congress on Computational Intelligence (IEEE WCCI 2024) to be held at Pacifico Yokohama, Yokohama, Japan, 30 June – 5 July 2024.

The field of imbalanced self-supervised learning, especially in the context of tabular data, has not been extensively studied. Existing research has predominantly focused on image datasets. This paper aims to fill this gap by examining the specific challenges posed by data imbalance in self-supervised learning in the domain of tabular data, with a primary focus on autoencoders. Autoencoders are widely employed for learning and constructing a new representation of a dataset, particularly for dimensionality reduction. They are also often used for generative model learning, as seen in variational autoencoders. When dealing with mixed tabular data, qualitative variables are often encoded using a one-hot encoder with a standard loss function (MSE or Cross Entropy). In this paper, we analyze the drawbacks of this approach, especially when categorical variables are imbalanced. We propose a novel metric to balance learning: a Multi-Supervised Balanced MSE. This approach reduces the reconstruction error by balancing the influence of variables. Finally, we empirically demonstrate that this new metric, compared to the standard MSE: i) outperforms when the dataset is imbalanced, especially when the learning process is insufficient, and ii) provides similar results in the opposite case.

Chapman & Hall/CRC Series in Actuarial Science

I am now part of the editorial team in charge of the Chapman & Hall/CRC Series in Actuarial Science, with Montserrat Guillén, Steven Haberman, Edward W. Frees (Jed) and Greg Taylor

The series provides a comprehensive and cutting-edge resource for professionals, academics, and students in the field of actuarial science. It aims to foster the exchange of knowledge, promote innovative research, and enhance the understanding of actuarial principles and practices. Contributions to the series will support one or more of the following key objectives.

  • Broad Coverage: This series covers a wide range of topics within actuarial science, including but not limited to life insurance, property and casualty (general) insurance, pension and retirement systems, health, and risk management.
  • Interdisciplinary Approach: We encourage contributions that bridge actuarial science with related fields such as statistics and probability, demography, finance and financial mathematics, insurance, economics, as well as data science, fostering a multidisciplinary perspective
  • Practical Insights: The series emphasizes practical applications, providing readers with real-world case studies, data analysis techniques, and modeling approaches used in actuarial practice.
  • Theoretical Advancements: We welcome contributions that advance the theoretical foundations of actuarial science, pushing the boundaries of knowledge and exploring emerging trends
  • Educational Resources: Our aim is to support actuarial education by offering textbooks that cater to both undergraduate and graduate students as well as professional education syllabi.
  • Global Perspective: We promote a global outlook by featuring works that address actuarial challenges and developments from diverse geographical regions and regulatory environments.
  • Innovation and Technology: As actuarial science evolves, we welcome works that explore the impact of emerging technologies on the profession, such as artificial intelligence and big data analytics.

The series is committed to upholding the highest standards of quality, rigor, and relevance. It seeks to be a valuable resource for practitioners seeking to enhance their skills, educators looking for comprehensive teaching materials, and researchers aiming to contribute to the advancement of actuarial science.