ItвЂ™s possible to fill in these blind spots through focussed effort. In 2017, a group of computer scientists at the University of California, Berkeley, pointed out that, on the Internet, almost all the images tagged as вЂњbedroomsвЂќ are вЂњclearly staged and depict a made bed from 2-3 meters away.вЂќ As a result, networks have trouble recognizing real bedrooms. Self-driving cars have been known to struggle with unusual signage, such as the blue stop signs found in Hawaii, or signs obscured by dirt or trees. Google asks users to label objects that its A.I.s are still learning to identify: thatвЂ™s what youвЂ™re doing when you take those вЂњAre you a bot?вЂќ tests, in which you select all the squares containing bridges, crosswalks, or streetlights. In recent years, weвЂ™ve all joined in the effort: todayвЂ™s facial recognition is particularly good because people tag themselves in pictures that they upload to social networks. Deep-learning researchers, therefore, have learned to collect and label data on an industrial scale. A deep-learning system for recognizing faces might have to be trained on tens of thousands of portraits, and it wonвЂ™t recognize a dress unless itвЂ™s also been shown thousands of dresses. The drawback of deep learning is that it requires large amounts of specialized data. system youвЂ™ve heard of-Siri, AlphaGo, Google Translate-depends on the technique. AlexNet showed that the technique could be used to solve real-world problems, while still running quickly on cheap computers. (Eventually, the interior layers of such networks can come to resemble the human visual cortex: early layers detect simple features, like edges, while later layers perform more complex tasks, such as picking out shapes.) Deep learning had been around for years, but was thought impractical. As the network was shown new images, it guessed what was in them inevitably, it was wrong, but after each guess it was made to adjust the connections between its layers of neurons, until it learned to output a label matching the one that researchers provided. AlexNet was a neural network, вЂњdeepвЂќ because its simulated neurons were arranged in many layers. The prospect of seeing like a human was a distant dream.ДЄll this changed in 2012, when Alex Krizhevsky, a graduate student in computer science, released AlexNet, a program that approached image recognition using a technique called deep learning. The field of вЂњcomputer visionвЂќ was a zoo of algorithms that made do in the meantime. researchers used to think that, without some kind of model of how the world worked and all that was in it, a computer might never be able to distinguish the parts of complex scenes.

A computer has never untangled a cable, doesnвЂ™t get that glass is reflective, doesnвЂ™t know that trees sway in the wind. But, to a computer, itвЂ™s a mess of color and brightness and shadow. I canвЂ™t help but parse this scene-about a third of the neurons in my cerebral cortex are implicated in processing visual information. As I write this, I can look up at my shelves: they contain books, and a skein of yarn, and a tangled cable, all inside a cabinet whose glass enclosure is reflecting leaves in the trees outside my window. For decades, image recognition was one of the grand challenges in artificial intelligence. Now it appeared that we were in this world already, and that the frontier was pastry.Д¬omputers learned to see only recently. Eventually, the tech would get to the point where one could do such things routinely.

I had often imagined that, someday, IвЂ™d be able to point my smartphone camera at a peculiar flower and have it identified, or at a chess board, to study the position. The bakery system had the flavor of magic-a feat seemingly beyond the possible, made to look inevitable. I tried to gather myself while the attendant wrapped and bagged my items. The system had apparently recognized my pastries by sight. I watched as a set of jagged, neon-green squiggles appeared around each item, accompanied by its name in Japanese and a price. A nearby screen displayed an image, shot from above, of my doughnuts and Danish. At the urging of an attendant, I slid my items onto a glowing rectangle on the counter. But I was most surprised when I got to the register. What first struck me was the selection, which seemed endless: there were croissants, turnovers, Danishes, pies, cakes, and open-faced sandwiches piled up everywhere, sometimes in dozens of varieties. After taking a tray and tongs at the front, you browsed, plucking what you liked from heaps of baked goods. One morning in the spring of 2019, I entered a pastry shop in the Ueno train station, in Tokyo.

Bear claw pastry