This article was initially written in French and published in November 2021.
Rubinstein (2012) claimed that “in economic theory, like Harry Potter, the Emperor’s New Clothes, or King Solomon’s Tales, we play in imaginary worlds. Economic theory invents tales and calls them models. An economic model is also somewhere between fantasy and reality (…) The word ‘model’ sounds more scientific than the word ‘fable’ or ‘tale’, but I think we are talking about the same thing“. Today, very often, learning models will build a model, based on learning data, and the actuary’s job will be to make sense of it, to find the story – the fable – that it is possible to tell. Continue reading The myth of interpretability and explicability of models
This article was co-authored with Laurence Barry, and initially written in French during the 2020 Summer.
The early days of the SARS-CoV-2 (or COVID-19) pandemic have seen a proliferation of calls for “individual responsibility”, starting with strong calls (and even an obligation in some countries, including France) to stay home as much as possible in the early spring of 2020, before it became mandatory to wear a mask in public (often closed) places during the summer. To paraphrase Coluche « dire qu’il suffirait que les gens restent chez eux pour qu’on puisse sortir… ». This call for each person’s responsibility is made in the name of all and for the good of all, symbolizing this very particular solidarity that the pandemic reminds us of: the risk that I choose to run does not only concern my person but also constitutes a risk for those around me. To formulate it in probabilistic terms, McKendrick (1926) stated that “the probability of occurrence increases with the number of existing cases“. This conception of individual responsibility, which is quite intuitive a priori, actually runs counter to the classical conception of economics: the rational (and responsible) individual makes choices that concern him, and that concern only him. The collective good is deduced by summing up individual utilities, independent of each other. But here is the problem: with the epidemic, an interdependence of utilities is created, so that the well-being of so-and-so, who chooses not to wear a mask, can harm the health and therefore the utility of many other people. How then can we think in economic terms of this “individual responsibility” in the context of the epidemic?
Continue reading Reconciling collective risks and individual decisions
Vendredi, je participerais avec Rodolphe Bigot à la journée d’étude “Genre, Algorithmes et Droit” à Aix-en-Provence.
Nice review of our paper, with Vincent Grari, on montrealethics.ai.
Sacrificing predictive performance is often viewed as an unacceptable option in machine learning. However, we note that to satisfy a fairness objective, the predictive performance can be reduced too much, especially for generic fair algorithms. Therefore, we have developed a more suitable and practical framework by using autoencoders techniques.
A fair pricing model via adversarial learning
For Carl Sagan, “if you understand exponentials, the key to many of the secrets of the universe is in your hand”. But not everyone seems ready to unlock the secrets of the universe. Thus, in mid-November 2021, more than 18 months after the beginning of the SARS-COVID 19 pandemic, the Minister of Health stated “the circulation of the virus has accelerated for a few weeks now, with an increase of 30% to 40% per week. We are not yet in a so-called exponential phase” (quoted in Ouest France (2021)). Since an increase at a constant rate is precisely the definition of “exponential growth”, one may wonder about this statement, which reveals either a lack of numeracy on the part of our leaders, or an element of language, with the word “exponential” becoming a taboo word that should not be mentioned?
(this is the English tranlation of a post published in January 2022)
Continue reading The taboo of the exponential
Two years after the beginning of the COVID-19 pandemic, due to the SARS-CoV-2 virus, we have all become (pseudo)-experts in contagion models. But beyond diseases, these models based on networks of interactions between people are also commonly used to describe the spread of a computer virus, of social norms, of an idea or a rumor in a society, or of an economic crisis, as Kucharski (2020) reminds us.
(this is the English tranlation of a post published in April 2022)
Continue reading Modeling contagion
Zylberman (2021) noted that “this pandemic began with the first case, but it will not end with the last case (…) one cannot date the end of a pandemic and the beginning of an endemic. Yet, in mid-April, the French president slipped in an interview (Garnier (2022)) that “society [is] on its way out of COVID” implying that the SARS-CoV-2 pandemic was over. At the same time, the virus was still killing more than 100 people a day, according to official statistics. While it is legitimate to question what exactly is a “COVID death“, it may seem surprising that 100 deaths per day (for more than three consecutive months) were met with such indifference, and that such a level was interpreted as the end of the pandemic.
(this is the English translation of a blog post published a few days ago)
Continue reading Are there acceptable deaths? or how to end a pandemic
Zylberman (2021) notait « cette pandémie a commencé avec le premier cas, mais elle ne se terminera pas avec le dernier cas (…) on ne peut pas dater la fin d’une pandémie et le début d’une endémie ». Pourtant, mi-avril, le président français a glissé dans une entrevue (Garnier (2022)) que « la société [est] en sortie de COVID » laissant entendre que la pandémie de SARS-CoV-2 était terminée. À la même époque, le virus tuait encore plus d’une centaine de personnes par jour, selon les statistiques officielles. S’il est légitime de s’interroger sur ce qu’est précisément un « mort de la COVID », il peut sembler étonnant que 100 morts par jour (pendant plus de 3 mois consécutifs) aient suscité autant d’indifférence, et qu’un tel niveau soit interprété comme la fin de la pandémie.
Continue reading Y-a-t-il des morts acceptables ? ou comment finir une pandémie
Ce matin, je donnerais un court exposé sur le thème de l’équité (algorithmique) en assurance dans un webinaire organisé par IVADO et OBVIA. Les slides sont en ligne.
Je participerai jeudi soir (heure française) ou jeudi midi (heure de la côte est canadienne) à un évènement intitulé Data & Risques Climatiques, en visio. Les slides sont maintenant en ligne.