Researchers at the Massachusetts Institute of Technology (MIT) and the University of California, Berkeley have developed a technique for producing artificial intelligence (AI) inference algorithms that can generate explanations for data and calculate their accuracy.
The sequential Monte Carlo with probabilistic program proposals (SMCP3) method enables any probabilistic program to intelligently guess explanations of data.
The researchers demonstrated SMCP3's ability to enhance AI systems' accuracy for tracking three-dimensional objects and analyzing data, and to improve the accuracy of the algorithms' own estimates of the data's qualtiy.
MIT's George Matheos said, "With SMCP3, I think it will be possible to use many more of these smart but hard-to-trust algorithms to build algorithms that are uncertainty-calibrated."
From MIT News
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, DC, USA
No entries found