Sigh. The phrase "neural network" is getting tossed around these days with some type of sensationalist flair, an almost romanticized notion of this impending explosion of super-human phenomena. As Alex Smola says with much frustration in one his classes, "it's only math!" It's a fancy term for straightforward mathematics. Bloggers are so often making them out to be much more than they are.
Sigh. The phrase "neural network" is getting tossed around these days with some type of sensationalist flair, an almost romanticized notion of this impending explosion of super-human phenomena.
It's all cyclical. This is at least the second, if not the third, wave of hype for Neural Networks. I remember a period back in the mid to late 90's when this stuff was quite the rage.
There's something to be said for both sides of this. I think you're right about the way NN's have been blown out of proportion. I find as I work across fields they're the most abused and misunderstood ML technique, since everyone and their dog has heard of NN's at this point. But on the other hand, there's a lot that can be said for simple building blocks combining to form something very complex. The biological "inspiration" behind NN's make them an attractive starting point for investigating this line of thought.
