On: AI models that predict disease are not as accurate as reports might suggest

[nostromo]

I’m confused by your comment because these are exactly the type of problems that humans generally really do a poor job classifying.

[jeffreyrogers → nostromo]

Most modern ML techniques do a poor job on these types of problems too unless they have a lot of data (hence the reference to sparsity) or assume structure that requires domain specific modeling to capture.

[buscoquadnary → nostromo]

It could be that after we train a biological nural net for decades it can get pretty good at intuiting things even if it can't explain how.

The numeral net in question is the Drs. Brain.