Watch this mashup of clips of cinema's most memorable "unfeeling" machines and you'll quickly get the idea that our greatest fear of artificial intelligence is that will treat us just like people do.

In a future in which machines approximate human intelligence, will the ability to fall in love become the new Turing test? Contra everyone's expectations, will the fickle hearts of humans mean robots are the ones left feeling unrequited?

This video is a product of a collaboration between filmmaker John Pavlus and musician Jascha Hoffman. The full album is available for free, for now.

If you're like me, you used to rely on your friends to tell you which movies were good and which books were worth reading. But now if you talk about such things at all, it's to lament the length of your Netflix queue or Amazon wishlist. In other words, to declare your free time unequal to the tasks put before it by the sophisticated recommendation engines employed by the internet's cybernetic arbiters of taste.

If there's one area from which the last vestiges of human interaction need to be expunged in the name of profit, it's music. AmIRight? Music discovery, after all, is littered with media that aggregate the opinions of insufferable young people. (Blog names like My Band Is Better Than Your Band pretty much sum it up.) These are then meta-aggregated by services like Hype Machine. (Which, god willing, will finally put Pitchfork out of its misery.)

But this just isn't enough for finnicky consumers of music. Maybe it's that music is so personal, so accessible -- so vast. A study commissioned by Orpheus Media Research, makers of music discovery software Myna, declares that "the accuracy of available [music] recommendation tools is lacking, with 40 percent [of survey respondents] saying that the results are accurate 50 percent or less of the time."

Which means most users of services like Pandora are happy customers. But in the age of the Internet there's always room for optimization -- people will pay scalpers up to $2000 for a marginally lighter and faster iPad, after all.

That's where Myna music comes in. I've been slagging their survey up to this point, but all in good fun: the service itself is quite impressive. Built by a classically trained musician, it claims to be able to use both the raw characteristics of a track as well as human psychology -- aka music cognition -- to automatically determine traits of a song as ephemeral as mood.

It's a service that, like Pandora, uses music to find music. But as anyone who has used Pandora can tell you, the service can feel a little limited at times -- bound by the conventions of genre and devoid of those serendipitous moments of musical connection that characterize the best human-generated mixes. I for one can't help wonder if this is a product of Pandora's human-dependent system, in which music is matched by tags determined assigned by trained musicians.

If Myna can transcend the limits of its human-augmented predecessors it will be no mean feat. Unfortunately, a consumer-facing version of the service has yet to go live. In the meantime, check out this demo, which looks promising:

Follow Mims on Twitter or contact him via email.

New work from students at the University of Hong Kong describes a novel use of neural networks, collections of artificial neurons or nodes that can be trained to accomplish a wide variety of tasks, previously used only in image recognition. The students used a convolutional network to "learn" features, such as tempo and harmony, from a database of songs that spread across 10 genres. The result was a set of trained neural networks that could correctly identify the genre of a song, which in computer science is considered a very hard problem, with greater than 87 percent accuracy. In March the group won an award for best paper at the International Multiconference of Engineers and Computer Scientists.

What made this feat possible was the depth of the student's convolutional neural network. Conventional "kernel machine" neural networks are, as Yoshua Bengio of the University of Montreal has put it, shallow. These networks have too few layers of nodes--analogous to the layers of neurons in your cerebral cortex--to extract useful amounts of information from complex natural patterns.

In their experiments, the students, led by professor Tom Li, discovered that the optimal number of layers for musical genre recognition was three convolutional (or "thinking") layers, with the first layer taking in the raw input data and the third layer outputting the genre data.

In each layer (pictured above) a single node, or neuron, "hears" only a tiny portion of the song, about 23 milliseconds. Each node overlaps 50 percent with its neighbors, however, and so in total the many nodes in the neural network hear a little more than two seconds of the song.

While a human might be hard-pressed to identify the genre of a track in so short a time, this particular algorithm does so easily when applied to songs from the standard library used for testing automated genre recognition. However, it fell flat in subsequent tests in which the students exposed it to music outside of the library on which it was trained.

They attribute the failure of their algorithm to work "in the wild" to an insufficiently large training library on which the network learned in the first place. Because their algorithm was able to chew through 240 songs in just two hours, the Hong Kong students say it has the potential to be quite scalable.

Intriguingly, the convoluted neural network on which this work is based was originally inspired by an examination of the cat visual cortex. Cats, being mammals, have visual cortexes not unlike our own. Experiments done in a related species, the ferret, have shown that, in the inverse of what was done in this paper where a visual neural network was applied to a problem in hearing, it's possible to re-wire a mammalian brain to see with its auditory cortex.

If convoluted neural networks are as flexible as the perceptual systems of mammals on which they are based, why aren't they being applied to all sorts of other problems of perception in AI?