Comment on A new link to an old model could crack the mystery of deep learningparentComments−ravi-delia4yI think you conflate several ideas of understanding here. We don't have good ways of predicting chaotic systems, but we often know a lot about them. Just knowing they're highly sensitive would be hard to confirm without math.
Comments
I think you conflate several ideas of understanding here. We don't have good ways of predicting chaotic systems, but we often know a lot about them. Just knowing they're highly sensitive would be hard to confirm without math.