There is a risk - and I see it instantiated in this post - of confusing two concepts with the label 'deep learning'. The one, typified by the chart at the top of the post, focuses on the distinction between understanding and mere memorization. The other, typified by the network diagram below, refers to unsupervised learning in neural networks - that is, learning that occurs without a 'training set' of previously resolved phenomena. We can learn from one about the other. But it is important not to conflate these distinct meanings.
Today: 7 Total: 90 [Direct link] [Share]
