[Statlist] Research Webinar in Statistics *FRIDAY, 16 OCTOBER 2020* GSEM, University of Geneva

[gsem-support-instituts]

Dear All,

We are pleased to invite you to our next Research webinar.

Looking forward to seeing you


Organizers :                                                                                   
E. Cantoni - S. Engelke - D. La Vecchia - E. Ronchetti
S. Sperlich - F. Trojani - M.-P. Victoria-Feser


FRIDAY, 15 OCTOBER 2020 at 11:15am
ONLINE
Please join the Zoom research webinar: https://unige.zoom.us/j/99238951053?pwd=dkd5UlRlYXkvNzZicnY0UlBCeW5rdz09
Password: 419459


Bayesian Optimal Experimental Design for Group Testing
(https://ai.googleblog.com/2020/07/exploring-faster-screening-with-fewer.html?m=1)
Marco CUTURI (https://marcocuturi.net/) - Google Brain & CREST - ENSAE, France

ABSTRACT:
In this talk, I will present recent work carried out to improve the performance of group testing using Bayesian analysis. The idea of group testing is that one can do better that testing people individually for a rare disease by pooling samples, and detecting whether one or possibly more people in a group are positive with a single test. We show in this work that a combination of sequential monte carlo with greedy combinatorial optimizers yields result that beat state of the art approaches.


Visit the website: https://www.unige.ch/gsem/en/research/seminars/rcs/