On: The Limitations of Deep Learning

[honestoHeminway]

Much of the programming with ML has moved towards cleaning, extrapolating and generating the data.

But this type of programing is - miracles- bugfree. We never hear of data-conversion gone wrong, data corrupted or data-mining withou conclusive results here. Obviously such bugs lack the glamour of security bugs.

[skgoa → honestoHeminway]

It's also very difficult to catch these errors. Your trained model just doesn't work as well as it could, but how would you be able to tell?