OPOMNIK: IJS KOLOKVIJ, sreda, 14. 4. 2021, ob 13. uri, dr. Mitja Luštrek

Tue Apr 13 10:13:15 CEST 2021

Vabimo vas na 3. predavanje iz sklopa "Kolokviji na IJS" v letu 2020/21, ki bo 14. aprila 2021, ob 13. uri. Predavanje bo v spletnem konferenčnem sistemu Zoom in mu lahko prisostvujete tako, da v brskalnik prenesete naslov  <https://uni-lj-si.zoom.us/j/98099345188?pwd=c1M0dHdnNHVDVFc2aC90RktCUHJKUT09> https://uni-lj-si.zoom.us/j/98099345188?pwd=c1M0dHdnNHVDVFc2aC90RktCUHJKUT09; identifikacijska stevilka sestanka je 980 9934 5188, geslo pa 876849.

Predavanje lahko spremljate tudi na TV IJS kanalu na spletnem naslovu tv.ijs.si <http://tv.ijs.si> .

Napovednik predavanja najdete na naslovu  <http://www.ijs.si/ijsw/Koledar_prireditev> http://www.ijs.si/ijsw/Koledar_prireditev, posnetke preteklih predavanj pa na  <http://videolectures.net/kolokviji_ijs> http://videolectures.net/kolokviji_ijs.

dr. Mitja Luštrek

Institut Jožef Stefan

Optimizacija ukrepov za zajezitev epidemije COVID-19

Organizacija XPrize prireja prestižna tekmovanja, ki spodbujajo razvoj tehnologij za reševanje pomembnih problemov človeštva. V XPrizovem Izzivu odziva na pandemijo (XPrize Pandemic Response Challenge) je bila naloga tekmovalcev napovedati, kako se okužbe s COVID-19 odzivajo na protiukrepe (kot so denimo zapiranje različnih ustanov in uporaba mask), ter predlagati učinkovite načrte protiukrepov za različne epidemiološke situacije. V predavanju bomo predstavili rešitev ekipe Odseka za inteligentne sistem Instituta “Jožef Stefan”, ki je na tekmovanju dosegla drugo mesto. Naša rešitev okužbe napoveduje s kombinacijo klasičnega epidemiološkega modela in strojnega učenja. Nato protiukrepe s čim boljšimi kompromisi med učinkom na okužbe in družbeno-ekonomsko ceno poišče s pomočjo algoritmov, ki se zgledujejo po biološki evoluciji.

Predavanje bo v angleščini.

*******

We invite you to the 3rd Institute colloquium in the academic year 2020/21. The colloquium will be held on Tuesday April 14, 2020 at 1 PM. The colloquium will be organized online using the zoom conference platform. To attend the colloquium, please copy the  <https://uni-lj-si.zoom.us/j/98099345188?pwd=c1M0dHdnNHVDVFc2aC90RktCUHJKUT09> https://uni-lj-si.zoom.us/j/98099345188?pwd=c1M0dHdnNHVDVFc2aC90RktCUHJKUT09; meeting ID is 980 9934 5188, password is 876849.

The colloquium is also aired on TV IJS at tv.ijs.si <http://tv.ijs.si> .

To read the abstract click  http://www.ijs.si/ijsw/Koledar_prireditev. Past colloquia are posted on  http://videolectures.net/kolokviji_ijs.

dr. Mitja Luštrek

Jožef Stefan Institute

Optimizing interventions against COVID-19

XPrize Foundation organizes high-profile competitions to develop technologies that solve the world's grand challenges. The competitors of the XPrize Pandemic Response Challenge were tasked with predicting how COVID-19 infections respond to various interventions (such as lockdowns and mask usage), and to propose effective plans of such interventions for different epidemiological situations. In this talk we will describe the solution developed by the team from the Department of Intelligent Systems at Jožef Stefan Institute, which placed second in the competition. The solution combined a classical epidemiological model with machine learning to predict future infections. Then it used algorithms inspired by biological evolution to find interventions plans with optimal trade-offs between the impact on the infections and the socio-economic cost.

Cordially invited!

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.ijs.si/pipermail/kolokvij-ijs/attachments/20210413/b747d8c8/attachment.htm>