Systems and methods for accelerating hessian-free optimization for deep neural networks by implicit preconditioning and sampling

Number of patents in Portfolio can not be more than 2000

United States of America Patent

PATENT NO 10056075
APP PUB NO 20170092263A1
SERIAL NO

15373775

Stats

ATTORNEY / AGENT: (SPONSORED)

Importance

Loading Importance Indicators... loading....

Abstract

See full text

A method for training a deep neural network, comprises receiving and formatting speech data for the training, preconditioning a system of equations to be used for analyzing the speech data in connection with the training by using a non-fixed point quasi-Newton preconditioning scheme, and employing flexible Krylov subspace solvers in response to variations in the preconditioning scheme for different iterations of the training.

Loading the Abstract Image... loading....

First Claim

See full text

Family

Loading Family data... loading....

Patent Owner(s)

  • INTERNATIONAL BUSINESS MACHINE CORPORATION

International Classification(s)

  • [Classification Symbol]
  • [Patents Count]

Inventor(s)

Inventor Name Address # of filed Patents Total Citations
Horesh, Lior North Salem, US 109 254
Kingsbury, Brian E D Cortlandt Manor, US 29 493
Sainath, Tara N Jersey City, US 163 2855

Cited Art Landscape

Load Citation

Patent Citation Ranking

Forward Cite Landscape

Load Citation

Maintenance Fees

Fee Large entity fee small entity fee micro entity fee due date
7.5 Year Payment $3600.00 $1800.00 $900.00 Feb 21, 2026
11.5 Year Payment $7400.00 $3700.00 $1850.00 Feb 21, 2030
Fee Large entity fee small entity fee micro entity fee
Surcharge - 7.5 year - Late payment within 6 months $160.00 $80.00 $40.00
Surcharge - 11.5 year - Late payment within 6 months $160.00 $80.00 $40.00
Surcharge after expiration - Late payment is unavoidable $700.00 $350.00 $175.00
Surcharge after expiration - Late payment is unintentional $1,640.00 $820.00 $410.00