Invention Grant
US08489528B2 Systems and methods for training neural networks based on concurrent use of current and recorded data
有权
基于同时使用当前和记录数据训练神经网络的系统和方法
- Patent Title: Systems and methods for training neural networks based on concurrent use of current and recorded data
- Patent Title (中): 基于同时使用当前和记录数据训练神经网络的系统和方法
-
Application No.: US12845567Application Date: 2010-07-28
-
Publication No.: US08489528B2Publication Date: 2013-07-16
- Inventor: Girish V. Chowdhary , Eric N. Johnson , Seung Min Oh
- Applicant: Girish V. Chowdhary , Eric N. Johnson , Seung Min Oh
- Applicant Address: US GA Atlanta
- Assignee: Georgia Tech Research Corporation
- Current Assignee: Georgia Tech Research Corporation
- Current Assignee Address: US GA Atlanta
- Agency: Troutman Sanders LLP
- Agent Jay R. Anderson, Esq.; Ryan A. Schneider, Esq.
- Main IPC: G06N3/08
- IPC: G06N3/08

Abstract:
Various embodiments of the invention are neural network adaptive control systems and methods configured to concurrently consider both recorded and current data, so that persistent excitation is not required. A neural network adaptive control system of the present invention can specifically select and record data that has as many linearly independent elements as the dimension of the basis of the uncertainty. Using this recorded data along with current data, the neural network adaptive control system can guarantee global exponential parameter convergence in adaptive parameter estimation problems. Other embodiments of the neural network adaptive control system are also disclosed.
Public/Granted literature
- US20110161267A1 SYSTEMS AND METHODS FOR TRAINING NEURAL NETWORKS BASED ON CONCURRENT USE OF CURRENT AND RECORDED DATA Public/Granted day:2011-06-30
Information query