Youyang Gu COVID model

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

The Youyang Gu COVID-19 model (sometimes abbreviated YYG[1]) is a computer software disease model for COVID-19 produced by independent data scientist Youyang Gu.[2]

The model is unique in applying machine learning to derive the basic reproduction number (R0) from data published by Johns Hopkins University's Center for Systems Science and Engineering (CSSE), and it seeks to minimize the error between its projections and CSSE data on the number of United States COVID-19 deaths.[3][4]

Use and endorsements[edit]

Gu's model was one of seven featured in The New York Times' survey of models and one of nine in FiveThirtyEight's survey,[5][6] was cited by the Centers for Disease Control (CDC) in its estimates for U.S. recovery,[7] and was one of three listed by the State of Washington on its "COVID-19 risk assessment dashboard" used to determine the date the state would reopen its economy after the COVID-19 pandemic in Washington.[8] The model's author claims it is the only one cited by CDC that is not receiving public funding.[9]

Yann LeCun, Facebook's chief AI scientist and professor at the Courant Institute of Mathematical Sciences, stated in May 2020 that Gu's model "is the most accurate to predict deaths from COVID-19", surpassing the accuracy of the well-funded Institute for Health Metrics and Evaluation COVID model.[10] Its superior accuracy was also noted by Silicon Valley newspaper The Mercury News[2] and by The Economist, which called it "more accurate than forecasts from many established outfits".[11]

Youyang Gu biography[edit]

Gu is a 2015 graduate of Massachusetts Institute of Technology.[12] He was born in Shanghai in 1993 or 1994 and grew up in Urbana, Illinois.[13]

See also[edit]

References[edit]

  1. ^ "COVID-19 forecasts". U.S. Centers for Disease Control. 11 February 2020. Retrieved 25 May 2020.
  2. ^ a b Vongs, Pueng (24 May 2020). "California prediction: AI-driven model tracks coronavirus infections, deaths". The Mercury News. San Jose, California.
  3. ^ Alper, Jarod. "WXML 2020 covid-modeling learning guide". Mathematics. Seattle, WA: University of Washington. Retrieved 25 May 2020.
  4. ^ Carlson, Joe; Howatt, Glenn (14 May 2020). "Health experts: Reopening businesses in Minnesota will spread COVID-19, but how much?". Star Tribune. Minneapolis, MN.
  5. ^ Bui, Quoctrung; Katz, Josh; Parlapiano, Alicia; Sanger-Katz, Margot (12 May 2020). "Coronavirus models are nearing consensus, but reopening could throw them off again". The New York Times.
  6. ^ Best, Ryan; Boice, Jay (May 2020). "Where the latest COVID-19 models think we're headed — and why they disagree". FiveThirtyEight.
  7. ^ Coleman, Justine (6 May 2020). "Scientist whose coronavirus model is used by CDC warns states may have to close again". The Hill.
  8. ^ "COVID-19 risk assessment dashboard". Joint Information Center. Washington State Emergency Operations Center. Retrieved 25 May 2020.
  9. ^ "Who we are: About covid19-projections.com". covid19-projections.com. Retrieved 25 May 2020.
  10. ^ LeCun, Yann [@ylecun] (18 May 2020). "Youyang Gu's model is the most accurate to predict deaths from Covid-19" (Tweet) – via Twitter.
  11. ^ "Early projections of COVID-19 in America underestimated its severity – by luck or by design, they have improved markedly since". The Economist. 23 May 2020.
  12. ^ Fottrell, Quentin (10 May 2020). "New estimates on coronavirus fatalities make for chilling reading as U.S. states ease restrictions on social distancing". MarketWatch.
  13. ^ Marek, Lynne (16 April 2015). "Chicago trading firms vie for top students". Crain's Chicago Business.

External links[edit]