RT @kph3k: The results of the Fragile Families Challenge are essential reading for anyone who has been following debates about polygenic sc…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
The results of the Fragile Families Challenge are essential reading for anyone who has been following debates about polygenic scores. Why? B/c our intuitions about PGS effect sizes need to be grounded in reality about our (in)ability to predict child outc
RT @eollion: What happens when machine learning does not corroborate the results from our classic (parametric regressions) tools? A great a…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
Amazing collection of perspectives on predicting life outcomes in complex data.
RT @eollion: What happens when machine learning does not corroborate the results from our classic (parametric regressions) tools? A great a…
RT @eollion: What happens when machine learning does not corroborate the results from our classic (parametric regressions) tools? A great a…
What happens when machine learning does not corroborate the results from our classic (parametric regressions) tools? A great and sobering experience.
Great thread
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
Thread summarizing all the research that came out of the Fragile Families Challenge 👇🏽
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @aselbst: This is pretty important work.
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @msalganik: Paper at https://t.co/WxDJbw0joz. Replication materials at https://t.co/GlhIDOWoNl.
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
RT @soizicpenicaud: Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!…
Etude américaine à lire sur les enjeux de prédiction dans les politiques publiques (aides sociales, justice). 160 (!!!) équipes de recherche ont essayé de prédire des trajectoires de vie avec du machine learning... **aucune** n’a obtenu de résultat satis
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @aselbst: This is pretty important work. https://t.co/YvwOfqfQ3c
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
Nice thread on predictive algorithms...
RT @pierre_bat: Un thread +++ quant à l'usage des modélisat° raffinées en sciences sociales -> 1- réunissez des équipes de pointe 2- don…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @pierre_bat: Un thread +++ quant à l'usage des modélisat° raffinées en sciences sociales -> 1- réunissez des équipes de pointe 2- don…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @aselbst: This is pretty important work.
What a powerful way to renind us of the current limits of predictive [social] anything.
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @pierre_bat: Un thread +++ quant à l'usage des modélisat° raffinées en sciences sociales -> 1- réunissez des équipes de pointe 2- don…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
So what we’re saying is that in this case data only helped us guess a little better than random. We are human before we are data points. #DataScience
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @Kelle_Parsons: So fascinating - a good reminder of the limits of predictive models on complex topics.
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @bfuste: Amics que sospireu per construir models predictius que puguin ser útils per a la definició de polítiques socials, això us inter…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @bfuste: Amics que sospireu per construir models predictius que puguin ser útils per a la definició de polítiques socials, això us inter…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
Un thread +++ quant à l'usage des modélisat° raffinées en sciences sociales -> 1- réunissez des équipes de pointe 2- donnez leur le même jeu de données 3- demandez leur de prévoir les résultats de la prochaine vague d'une grande enquête C'est ce qu
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @MDRC_News: Using #MachineLearning, hundreds of researchers tried to predict 6 life outcomes, such as a child’s GPA & whether a family w…
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
Measuring the predictability of life outcomes with a scientific mass collaboration https://t.co/uhilonfftf https://t.co/bcJwRFR0JQ
Amics que sospireu per construir models predictius que puguin ser útils per a la definició de polítiques socials, això us interessarà molt (i us animarà poc...)
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…
RT @IEthics: Thread #ethics #data #tech #algorithms #AI #gov #research https://t.co/StiAFB54NM
RT @msalganik: If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcome…