RT @MoritzBuchi: "In many cases, simpler models outperformed more complicated techniques, and teams with more accurate scoring models came…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
Measuring the predictability of life outcomes with a scientific mass collaboration | PNAS https://t.co/K4lag1lU26
"In many cases, simpler models outperformed more complicated techniques, and teams with more accurate scoring models came from uncommon disciplines" @msalganik https://t.co/OLd5WTEisv Paper: https://t.co/xol1LsY4SI https://t.co/Tx63803Wy2
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
"Hundreds of researchers attempted to predict six life outcomes... [using] machine-learning methods optimized for prediction... on a vast dataset... collected by social scientists over 15 y. However, no one made very accurate predictions. https://t.co/Wrf
RT @ChrisDanforth: Predicting life outcomes is hard: "Despite using a rich dataset & machine-learning methods... the best predictions were…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
“In the Fragile Families Challenge, the simple benchmark model with only a few predictors was only slightly worse than the most accurate submission, and it actually outperformed many of the submissions”
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
On the need of sociological understanding in the age of algorithmic prediction: "Despite using a rich dataset and applying machine-learning methods optimized for prediction, the best predictions were not very accurate" Great work
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
That damn randomness is raising its ugly head again... (aka, the gloomy prospect/unexplained non-shared environment)
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @stuartbuck1: Predictive modeling is hard! From a paper published today: https://t.co/m7YaUkf0uZ https://t.co/KxRBUvkzy5
RT @stuartbuck1: Predictive modeling is hard! From a paper published today: https://t.co/m7YaUkf0uZ https://t.co/KxRBUvkzy5
RT @syardi: this abstract 😮😬 "Hundreds of researchers attempted to predict six life outcomes... However, no one made very accurate predicti…
This study should be the watermark behind all AI/ML hype talks.
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @amaatouq: Such an honor to play a small part in this truly inspirational collaborative effort (led by @msalganik @IanLundberg1 @alextki…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
Fragile Families Challenge というデータ解析コンペをまとめた総勢113人の著者による論文が PNASに掲載されました。自分はいち参加者なので貢献は僅かですが、このプロジェクトに参加できたことを光栄に思います。オーガナイザーチーム、特にMattには大感謝です。 https://t.co/HzlAE8d9W9
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
Hugely important thread on the relative effectiveness of machine learning techniques to predict key social outcomes. Spoiler: not particularly effective.
RT @squazzoni: Brilliant study on life trajectories using predictive models by @msalganik and a huge team. Must read
Brilliant study on life trajectories using predictive models by @msalganik and a huge team. Must read
Hundreds of researchers attempted to predict six life outcomes using #MachineLearning with a rich dataset. No one made very accurate predictions. Also, good luck citing all the authors of this paper!
Measuring the predictability of life outcomes with a scientific mass collaboration via @PNASNews https://t.co/i4mNuaqU8w https://t.co/8B1KCUzee6
"before using complex predictive models, we recommend that policymakers determine whether the achievable level of predictive accuracy is appropriate for the setting where the predictions will be used" - @msalganik et al. (many al!) in @PNASNews https://t.c
RT @jblumenstock: Great article highlighting inherent difficulties in predicting life outcomes (relevant to ML in social science more broad…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
This study is amazing—and immensely personally validating (I'm sure it is for few others too)
RT @autismcrisis: "Hundreds of researchers attempted to predict six life outcomes" but "no one made very accurate predictions"? https://t.c…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @clary_co: "The best submissions, which often used complex machine-learning methods & had access to 1000s of predictor variables, were o…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
Predictive modeling is hard! From a paper published today: https://t.co/m7YaUkf0uZ https://t.co/KxRBUvkzy5
RT @amaatouq: Such an honor to play a small part in this truly inspirational collaborative effort (led by @msalganik @IanLundberg1 @alextki…
RT @hcp4715: "These researchers used machine-learning methods optimized for prediction, and they drew on a vast dataset that was painstakin…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @alondra: Important finding: “researchers attempted to predict six life outcomes...used machine-learning methods optimized for predictio…
RT @alondra: Important finding: “researchers attempted to predict six life outcomes...used machine-learning methods optimized for predictio…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @lpargyle: Pleased to be one of more than 100 contributors to this very cool, cutting edge computational social science project. Out tod…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
Pleased to be one of more than 100 contributors to this very cool, cutting edge computational social science project. Out today in PNAS.
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
Great article highlighting inherent difficulties in predicting life outcomes (relevant to ML in social science more broadly) https://t.co/tTMSJtvUgc #FragileFamiliesChallenge @msalganik @duncanjwatts @mrtz @random_walker @karen_ec_levy +many others!
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @jsradford: Reminds me of the debate @MishaTeplitskiy and I have about sociology's 'dark matter' - our persistently low r-squared. I sug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
Such an honor to play a small part in this truly inspirational collaborative effort (led by @msalganik @IanLundberg1 @alextkindel & S McLanahan). Can the outcomes of children from vulnerable families be predicted in a #kaggle-style competition? https:/
“... results suggest practical limits to the predictability of life outcomes in some settings...” https://t.co/YSWDvEPdky
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
Measuring the predictability of life outcomes with a scientific mass collaboration. A fantastic effort of social #research https://t.co/upWrDsGRUB
On one hand, more evidence that it's hard for data-driven models to outperform bespoke expert models. On the other hand, even bespoke models have a hard time predicting the future, suggesting that many outcomes we might care about are not that stable over
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @alondra: Important finding: “researchers attempted to predict six life outcomes...used machine-learning methods optimized for predictio…
Important finding: “researchers attempted to predict six life outcomes...used machine-learning methods optimized for prediction, and they drew on a vast dataset...However, no one made very accurate predictions...these results raise a number of concerns” fo
RT @jsradford: Reminds me of the debate @MishaTeplitskiy and I have about sociology's 'dark matter' - our persistently low r-squared. I sug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @Zoe_LeBlanc: this is fascinating!! CSS giant study on value of ML "only somewhat better than the results from a simple benchmark model…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
RT @duncanjwatts: Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Hug…
Proud to be associated with this breakthrough collaborative effort, unprecedented (I believe) in the social sciences. Huge kudos to @msalganik and Sarah McLanahan Measuring the predictability of life outcomes with a scientific mass collaboration https://t.
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
this abstract 😮😬 "Hundreds of researchers attempted to predict six life outcomes... However, no one made very accurate predictions. For policymakers considering using predictive models... these results raise a number of concerns." https://t.co/kL2aXr4uHF
Measuring the predictability of life outcomes with a scientific mass collaboration https://t.co/CGWahpkszQ
RT @jsradford: Reminds me of the debate @MishaTeplitskiy and I have about sociology's 'dark matter' - our persistently low r-squared. I sug…
@Robsync1
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @ChrisDanforth: Predicting life outcomes is hard: "Despite using a rich dataset & machine-learning methods... the best predictions were…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @jsradford: Reminds me of the debate @MishaTeplitskiy and I have about sociology's 'dark matter' - our persistently low r-squared. I sug…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
RT @chris_bail: 1/n Can machine learning solve social science problems? This unprecedented study led by @msalganik and dozens of co-authors…
such systems have been crimped or terminated for their un-PC findings, in the past - expect the same here -