Communications of the ACM
Machines That Learn Better
Cameron Freer, left, an MIT instructor in pure mathematics, and Daniel Roy, right, a Ph.D. student in the Department of Electrical Engineering and Computer Science, have described an inference algorithm that can handle a large class of problems involving
Credit: Jason Dorfman / CSAIL
In the last 20 years or so, many of the key advances in artificial-intelligence research have come courtesy of machine learning. A new approach called probabilistic programming makes it much easier to build machine-learning systems, but it's useful for a relatively narrow set of problems. Now, MIT researchers have discovered how to extend the approach to a much larger class of problems, with implications for subjects as diverse as cognitive science, financial analysis and epidemiology.
A handful of new, experimental, probabilistic programming languages—one of which, Church, was developed at MIT—promise to cut the time of building a machine-learning system down to a matter of hours, says Daniel Roy, a Ph.D. student in the Department of Electrical Engineering and Computer Science, who along with Cameron Freer, an instructor in pure mathematics, led the new research.
From MIT News
View Full Article
No entries found