Tweet by KordingLab 🦖 on Twitter error assess suicide

C-5SttwO_normal.png spacer.png
KordingLab 🦖
⁦‪@KordingLab‬⁩
logo_twitter-1497383721365.png
spacer_464x1-1582829598167.png
Machine learning can easily produce false positives when the test set is wrongly used. Just et al in ⁦‪@NatureHumBehav‬⁩ suggested that ML can identify suicidal ideation extremely well from fMRI and we were skeptical. Today retraction and our analysis of what went wrong came out.
4/6/23, 9:02 PM

Joseph Thornton

Tweet by Stanford HAI on Twitter. Apr 13

2gkqnAOS_normal.jpg spacer.png
Stanford HAI
⁦‪@StanfordHAI‬⁩
logo_twitter-1497383721365.png
spacer_464x1-1582829598167.png
On Apr 13, hear global perspectives on human-centered AI, including designing for underrepresented languages and developing AI tools to stop misinformation in global contexts. This event is co-sponsored with ⁦‪@StanfordGlobal‬⁩ ⁦‪@cas_stanford‬⁩ ⁦‪@csastanford‬⁩: stanford.io/3ZvCKMf
3/30/23, 3:01 PM

Joseph Thornton