Patent · US Active

Unsupervised learning utilizing sequential output statistics

US10776716B2 · kind B2 · utility

2Cited by

3References

20Claims

0Family size

Assignee

MICROSOFT TECHNOLOGY LICENSING, LLC · US

Inventors

Yu Liu · Austin, US
Jianshu Chen · Redmond, US
Li Deng · Redmond, US

Key dates

Filing date	Jun 13, 2017
Grant date	Sep 15, 2020
Priority date	—
Expiry date	May 25, 2039

Classification

Technology area (CPC G)Physics
CPC primaryG10L15/197
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

In classification tasks applicable to data that exhibit sequential output statistics, a classifier may be trained in an unsupervised manner based on a sequence of input samples and an unaligned sequence of output labels, using a cost function that measures the negative cross-entropy of an N-gram joint probability distribution derived from the sequence of output labels with respect to an expected N-gram frequency in a second sequence of output labels predicted by the classifier. In some embodiments, a primal-dual reformulation of the cost function is employed to facilitate optimization.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.