BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CREST - ECPv5.1.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:CREST
X-ORIGINAL-URL:https://crest.science
X-WR-CALDESC:Events for CREST
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20240101T000000
END:STANDARD
TZID:Europe/Helsinki
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:EEST
DTSTART:20240331T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:EET
DTSTART:20241027T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20240129T090000
DTEND;TZID=UTC:20240205T121500
DTSTAMP:20260711T081621
CREATED:20231019T102431Z
LAST-MODIFIED:20231019T102431Z
UID:16122-1706518800-1707135300@crest.science
SUMMARY:Biases\, Discrimination\, and Fairness\, Arthur Charpentier (Université du Québec à Montréal)
DESCRIPTION:  \n\n\n\n  \n  \nSCHEDULE\n  \nMonday\n  \n29th January 2024 \n5th February 2024\n  \nFrom 9:00 to 12:15\n  \nRoom 2033\n\n\n  \nThursday\n  \n1st February 2024\n  \nFrom 9:00 to 12:15\n  \nRoom 2033\n\n\n\nAims and objectives\nThis course will provide a state-of-the-art\, on fairness and discrimination\, in the context of insurance pricing (and more generally\, predictive models). As explained by Avraham et al. (2014) “‘insurance companies are in the business of discrimination. Insurers attempt to segregate insureds into separate risk pools based on the differences in their risk profiles\, first\, so that different premiums can be charged to the different groups based on their differing risks and\, second\, to incentivize risk reduction by insureds. This is why we let insurers discriminate. There are limits\, however\, to the types of discrimination that are permissible for insurers. But what exactly are those limits and how are they justified?”. First\, we will come back to the specificities of predictive models in insurance. We will come back to the different places where a potential discrimination can intervene\, by insisting on the possible biases in the data\, in the models. We will present in particular the regulations in Europe and North America. In a second step\, we will see how to quantify a possible discrimination\, insisting on the main measures of “group-fairness”\, before discussing the individual approach\, in particular in relation with the causal approaches. Indeed\, the central question of discrimination is “would the price have been different if this person had been a man instead of a woman”. We will see how to build a counterfactual allowing to quantify a possible discrimination. Finally\, we will see how to correct a discrimination\, insisting on the in-processing (throught penalized models) and post-processing approaches (using optimal transport). \n
URL:https://crest.science/event/biases-discrimination-and-fairness-arthur-charpentier-universite-du-quebec-a-montreal/
LOCATION:2033
CATEGORIES:Actuarial Science
ORGANIZER;CN="Christian-Yann%20Robert":MAILTO:christian-yann.robert@ensae.fr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20240129T121500
DTEND;TZID=Europe/Helsinki:20240129T133000
DTSTAMP:20260711T081621
CREATED:20240104T092642Z
LAST-MODIFIED:20240104T093349Z
UID:16410-1706530500-1706535000@crest.science
SUMMARY:Gökce Gökkoca (TSE) "Antibiotic Stewardship in Primary Care: Evidence from Pay-for-Performance in France"
DESCRIPTION:Applied Seminar \nTime: 12:15 pm – 13:30 pm\nDate: 29th of January\nRoom : 3001 \n  \nGökce Gökkoca (TSE) “Antibiotic Stewardship in Primary Care: Evidence from Pay-for-Performance in France” \nAbstract :This paper investigates whether financial incentives for curbing antibiotic prescriptions are effective and how the design of incentives plays a role in influencing physician behavior. Using prescription-level data from French general practitioners over six years\, I provide evidence of the incentives’ effectiveness by exploiting variation in the set of diseases that the physicians treat as well as in the reward scheme. To understand how they respond\, I propose a model that incorporates financial incentives into physician’s decision-making and test the predictions of the model. The results highlight that the reduction in antibiotic prescriptions varies across different diseases\, in line with the physicians’ altruism and\, hence\, the patient’s needs. Moreover\, forward-looking physicians are influenced by the marginal cost of antibiotic prescriptions and the design of the incentives. While the program is effective\, the magnitude is moderate\, with a 2 percentage point drop in the antibiotic prescription rate. Comparing the effect to the cost of the program\, conditioning the rewards on prescription rates rather than the improvement over time plays a role. As a result\, while aggregate bonus payments per physician remain modest (on average 0.2% of physicians’ annual income)\, the cost per avoided prescription is substantial (on average 56% of the fixed visit fee). \nOrganizer: Michael VISSER \n
URL:https://crest.science/event/gokce-gokkoca-tse-antibiotic-stewardship-in-primary-care-evidence-from-pay-for-performance-in-france/
CATEGORIES:Applied Seminar,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20240129T140000
DTEND;TZID=Europe/Helsinki:20240129T151500
DTSTAMP:20260711T081621
CREATED:20231222T093753Z
LAST-MODIFIED:20240118T150210Z
UID:16387-1706536800-1706541300@crest.science
SUMMARY:Gabriel PEYRE (Ecole Normale Supérieure) - Conservation Laws for Gradient Flows
DESCRIPTION:Statistical Seminar: Every Monday at 2:00 pm.\nTime: 2:00 pm – 3:15 pm\nDate: 29th January of 2024\nPlace : 3001 \n  \nGabriel PEYRE (Ecole Normale Supérieure) – Conservation Laws for Gradient Flows \n  \nAbstract: \nUnderstanding the geometric properties of gradient descent dynamics is a key ingredient in deciphering the recent success of very large machine learning models. A striking observation is that trained over-parameterized models retain some properties of the optimization initialization. This “implicit bias” is believed to be responsible for some favorable properties of the trained models and could explain their good generalization properties. In this talk I will first rigorously expose the definition and basic properties of “conservation laws”\, which are maximal sets of independent quantities conserved during gradient flows of a given model (e.g. of a ReLU network with a given architecture) with any training data and any loss. Then I will explain how to find the exact number of these quantities by performing finite-dimensional algebraic manipulations on the Lie algebra generated by the Jacobian of the model. In the specific case of linear and ReLu networks\, this procedure recovers the conservation laws known in the literature\, and prove that there are no other laws. \nThe associated paper can be found here https://arxiv.org/abs/2307.00144 and the open source code is here https://github.com/sibyllema/Conservation_laws.This is a joint work with Sibylle Marcotte and Rémi Gribonval. \n  \nOrganizers:\nCristina BUTUCEA (CREST)\, Anna KORBA (CREST)\, Karim LOUNICI (CMAP) \, Jaouad MOURTADA (CREST) \nSponsors:\nCREST-CMAP \n
URL:https://crest.science/event/gabriel-peyre-ecole-normale-superieure-to-be-announced/
CATEGORIES:Statistics
ATTACH;FMTTYPE=:
END:VEVENT
END:VCALENDAR