The changes in Google Translate illustrate how machine learning (of which deep learning is a subfield) has dramatically reduced the costs of quality-adjusted prediction. For the same cost in terms of computational capacity, Google can now provide higher-quality translations. The cost of producing the same quality of prediction has dropped significantly.
Innovations in prediction technology are having an impact on areas traditionally associated with forecasting, such as fraud detection. Credit card fraud detection has improved so much that credit card companies detect and address fraud before we notice anything amiss. Still, this improvement seems incremental. In the late 1990s, the leading methods caught about 80 percent of fraudulent transactions. These rates improved to 90–95 percent in 2000 and to 98–99.9 percent today. That last jump is a result of machine learning; the change from 98 percent to 99.9 percent has been transformational. The change from 98 percent to 99.9 percent might seem incremental, but small changes are meaningful if mistakes are costly. An improvement from 85 percent to 90 percent accuracy means that mistakes fall by one-third. An improvement from 98 percent to 99.9 percent means mistakes fall by a factor of twenty. An improvement of twenty no longer seems incremental.
The change from 98 percent to 99.9 percent might seem incremental, but small changes are meaningful if mistakes are costly. An improvement from 85 percent to 90 percent accuracy means that mistakes fall by one-third. An improvement from 98 percent to 99.9 percent means mistakes fall by a factor of twenty. An improvement of twenty no longer seems incremental.
The drop in the cost of prediction is transforming many human activities. Just as the first applications of computing applied to familiar arithmetic problems like census tabulations and ballistics tables, many of the first applications of inexpensive prediction from machine learning applied to classic prediction problems. In addition to fraud detection, these included creditworthiness, health insurance, and inventory management. Creditworthiness involved predicting the likelihood that someone would pay back a loan. Health insurance involved predicting how much an individual would spend on medical care. Inventory management involved predicting how many items would be in a warehouse on a given day.
More recently, entirely new classes of prediction problems emerged. Many were nearly impossible before the recent advances in machine intelligence technology, including object identification, language translation, and drug discovery. For example, the ImageNet Challenge is a high-profile annual contest to predict the name of an object in an image. Predicting the object in an image can be a difficult task, even for humans. The ImageNet data contains a thousand categories of objects, including many breeds of dog and other similar images. It can be difficult to tell the difference between a Tibetan mastiff and a Bernese mountain dog, or between a safe and a combination lock. Humans make mistakes around 5 percent of the time.
The current generation of AI is a long way from the intelligent machines of science fiction. Prediction does not get us HAL from 2001: A Space Odyssey, Skynet from The Terminator, or C3PO from Star Wars. If modern AI is just prediction, then why is there so much fuss? The reason is because prediction is such a foundational input. You might not realize it, but predictions are everywhere. Our businesses and our personal lives are riddled with predictions. Often our predictions are hidden as inputs into decision making. Better prediction means better information, which means better decision making.
This is an extract from Joshua Gans, Avi Goldfarb and Ajay Agrawal's Prediction Machines published by Harvard Business Review Press