Up in arms for what reason? That neural networks are perfectly interpretable? That's the nature of huge amorphous deep networks. These feature extraction forays

Up in arms for what reason? That neural networks are perfectly interpretable? That's the nature of huge amorphous deep networks. These feature extraction forays are a good step forward though.