id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
andromedayelton-com-9751		Of such stuff are (deep)dreams made: convolutional networks and neural style transfer – andromeda yelton			.html	text/html	1176	89	68	Of such stuff are (deep)dreams made: convolutional networks and neural style transfer – andromeda yelton The specifics vary by neural net and data set, but you might find that the first layer gets excited about straight lines and colors; the second about curves and simple textures (like stripes) that can be readily composed from straight lines; the third about complex textures and simple objects (e.g. wheels, which are honestly just fancy circles); and so on, until the final layers recognize complex whole objects. As you already expect if you have done a little machine learning, that means that we need to write cost functions that mean "how close is this image to the desired content?" and "how close is this image to the desired style?" And then there's a wrinkle that I haven't fully understood, which is that we don't actually evaluate these cost functions (necessarily) against the outputs of the neural net; we actually compare the activations of the neurons, as they react to different images — and not necessarily from the final layer!	./cache/andromedayelton-com-9751.html	./txt/andromedayelton-com-9751.txt