In my first memory of the medium, I am standing behind Eric and he has black hair. His mother couldn't have watched me more than a handful of times in my childhood, but I can visualise the living room vestibule where we are standing as clearly as if it's my own house. It must be around 1993. I remember that Eric is playing Super Mario Bros, but only because my little brother's name is Mario and I make the connection with confusion. Eric offers to let me play, once, and I hold the controller like the alien object it is.

The study states that deep learning's impressive progress has come with a "voracious appetite for computing power." Researchers at the Massachusetts Institute of Technology, MIT-IBM Watson AI Lab, Underwood International College, and the University of Brasilia have found that we are reaching computational limits for deep learning. The new study states that deep learning's progress has come with a "voracious appetite for computing power" and that continued development will require "dramatically" more computationally efficient methods. "We show deep learning is not computationally expensive by accident, but by design. The same flexibility that makes it excellent at modeling diverse phenomena and outperforming expert models also makes it dramatically more computationally expensive," the coauthors wrote.