Remember the recent @nature paper using deep learning #AI to predict aftershocks after an earthquake?
Well, this one under the heading "Matters Arising," says it ain't so, by @ArnaudMignan and Marco Broccardo
nature.com/articles/s4158…
![](https://pbs.twimg.com/media/EF5WE5LUYAAkH4t.jpg)
![](https://pbs.twimg.com/media/EF5WE5MUEAA5IGf.jpg)
nature.com/articles/s4158…
response from the original paper's author, along with original paper and @NatureNV nature.com/magazine-asset…
![](https://pbs.twimg.com/media/EF_Yx_4UEAE3Xnj.jpg)
![](https://pbs.twimg.com/media/EF_Yx_5U4AEZCZB.jpg)
![](https://pbs.twimg.com/media/EF_Yx_6VAAA-ydg.jpg)
![](https://pbs.twimg.com/media/EF_Yx_6UUAAEWRy.jpg)
"we demonstrate that simpler models seem to offer similar predictive powers, if not better ones."
by @ArnaudMignan and @MarcoBroccardo
arxiv.org/abs/1910.01178 @arxiv #AI
![](https://pbs.twimg.com/media/EGCjhn7U4AA_ZgB.jpg)