ImageNet Roulette

ImageNet is one of the most important and historically significant training sets in artificial intelligence. In the words of its creators, the idea behind ImageNet was to "map out the entire world of objects." After its initial launch in 2009, ImageNet grew enormous: the development team scraped a collection of many millions of images from the Internet and briefly became the world's largest academic user of Amazon's Mechanical Turk, using an army of piecemeal workers to sort an average of 50 images each minute into thousands of categories. When it was finished, ImageNet consisted of over 14 million labelled images organized into more than twenty thousand categories.