[blake_himself]

> When you are playing chess, at some point you make a mistake, you may go back several steps that were “correct” to find the one that was wrong. When you fall off a bicycle, you think of when you lost your balance. Deep learning does that. The credit assignment in a deep learning exercise can be tens, even hundreds, of levels deep.

This sounds like reinforcement learning. Anyone know what he's talking about, some RNN with 'tens, even hundreds' of levels of feedback that hasn't died away?

[jamiis]

Pretty sure he's just referring to backprop[0] correcting weights more dramatically at one layer vs. another layer

[0]http://en.m.wikipedia.org/wiki/Backpropagation