Here is an amazing post at the Google Research blog about some very interesting work being done in Neural Networks and image interpretation/classification.
I selected some quotes below – but the post is worth a read (emphasis added).
Layers of Abstraction
[T]he first layer maybe looks for edges or corners. Intermediate layers interpret the basic features to look for overall shapes or components, like a door or a leaf. The final few layers assemble those into complete interpretations—these neurons activate in response to very complex things such as entire buildings or trees.
Dumb Ideas About Dumbbells
This kind of result is important, since a system that emulates neural networks needs to be confused sometimes (it means that it is learning).
[I]t seems no picture of a dumbbell is complete without a muscular weightlifter there to lift them. In this case, the network failed to completely distill the essence of a dumbbell. Maybe it’s never been shown a dumbbell without an arm holding it.
Animals in Clouds: over-interpretation
The results are intriguing—even a relatively simple neural network can be used to over-interpret an image, just like as children we enjoyed watching clouds and interpreting the random shapes. This network was trained mostly on images of animals, so naturally it tends to interpret shapes as animals. But because the data is stored at such a high abstraction, the results are an interesting remix of these learned features.
Dreams and the Creative Process
If we apply the algorithm iteratively on its own outputs and apply some zooming after each iteration, we get an endless stream of new impressions, exploring the set of things the network knows about. We can even start this process from a random-noise image, so that the result becomes purely the result of the neural network.
[This] also makes us wonder whether neural networks could become a tool for artists—a new way to remix visual concepts—or perhaps even shed a little light on the roots of the creative process in general.
PS: more great images here.See more blog.