Artificial Intelligence will make better decisions by embracing uncertainty, which is the most powerful approach in AI, it is a way of gaining new capabilities.
Researchers at Uber and Google are working on modifications to the two most popular deep-learning frameworks that will enable them to handle probability. It will provide a way for the smartest AI programs to measure their confidence in a prediction or a decision – essentially, to know when they should doubt themselves.
Deep learning, which involves feeding example data to a large and powerful neural network, has been an enormous success over the past few years, enabling machines to recognize objects in images or transcribe speech almost perfectly. But it requires lots of training data and computing power, and it can be surprisingly brittle.
The new approach could be useful in critical scenarios involving self-driving cars and other autonomous machines.
"You would like a system that gives you a measure of how certain it is," says Dustin Tran, who is working on this problem at Google. "If a self-driving car doesn’t know its level of uncertainty, it can make a fatal error, and that can be catastrophic."
The work reflects the realization that uncertainty is a key aspect of human reasoning and intelligence. Adding it to an AI program could make it smarter and less prone to blunders, says Zoubin Ghahramani, a prominent AI researcher who is a professor at the University of Cambridge and chief scientist at Uber.
"We want to have a rock-solid framework for deep learning, but make it easier for people to represent uncertainty," Ghahramani said during a major AI conference in Long Beach, California.
Noah Goodman, a professor at Stanford who is also affiliated with Uber's AI Lab, explains that giving deep learning the ability to handle probability can make it smarter in several ways. It could, for example, help a program recognize things, with a reasonable degree of certainty, from just a few examples rather than many thousands.
And while a conventional deep-learning system learns only from the data it is fed, Pyro – a new programming language released by Uber that merges deep learning with probabilistic programming – can also be used to build a system preprogrammed with knowledge.
"In cases where you have prior-knowledge you want to build into the model, probabilistic programming is especially useful," Goodman says. "People will use Pyro for all sorts of things."
Edward – which is another programming language that embraces uncertainty – is developed by Columbia University and funded by DARPA. Both Pyro and Edward are still at early stages of development, but it is not hard to see why Uber and Google are interested.
Uber uses machine learning in countless areas, from routing drivers to self-driving cars. The company has invested heavily in AI, hiring a number of experts and working on new ideas. Google has rebuilt its entire business around AI and deep learning of late.
Pyro and Edward are significant for bringing together two competing schools in AI, one focused on neural networks and another on probability. "They can come together – in fact, they are coming together – in the tools that we are now building," adds Goodman.
Let us know what you think about the fact that AI is getting more human-like by the day.