CONSTRAINED LINE SEARCH OPTIMIZATION FOR DISCRIMINATIVE TRAINING OF HMMS

Patent №

US 8,239,332

Granted

2012-08-07

Filed 2007

Owner

MICROSOFT CORPORATION

AI components

3

ml · vision · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

11943210

An exemplary method for optimizing a continuous density hidden Markov model (CDHMM) includes imposing a constraint for discriminative training, approximating an objective function as a smooth function of CDHMM parameters and performing a constrained line search on the smoothed function to optimize values of the CDHMM parameters. Various other methods, devices and systems are disclosed.

AI classification

Machine learning1.00
AI hardware1.00
Vision0.52
Speech0.32
Natural language0.31
Evolutionary computation0.16
Knowledge representation0.14
Planning0.01

Ownership

MICROSOFT CORPORATION

assignment · 208670563

Assignors

LIU, PENG, JIANG, HUI, SOONG, FRANK KAOPING, LIU, PENG, JIANG, HUI, SOONG, FRANK KAO-PINGK

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC