How artificial neural networks build their own learning machinery

According to MIT News, the researchers argue in a paper published Oct. 9 in the journal PNAS[1] that a key test of whether a network learns in a biologically realistic way is its ability to organise itself "without the guiding hand of a developer."[2]MIT News reports that with some simple instructions about how and when connections should change, a group of randomly connected units can reshape itself into an effective information-processing network[1].Co-first author Qianli Liao told MIT News that "Current AI technology is working very well, but it doesn't necessarily use a learning algorithm that is used by the brain."[1]MIT News notes that most modern AI implements gradient descent through some form of the back-propagation algorithm, which neuroscientists say is unlikely to be used by the brain.[1] According to the outlet, the team compared its self-organizing network to networks designed to learn in this way and found that it performed at a similar level.[1]Professor Tomaso Poggio told MIT News that if the brain does use the proposed synaptic motifs, "then we will have a strong connection at the level of fundamental mechanisms between artificial and natural networks"[3].Mirage News, which republished the MIT story, reports that Liao says a self-organizing network does not necessarily learn the same way the brain does[3]. Tech Xplore reports that Liao points to the development of brain-computer interfaces[4] as one way the work could be used.