×

Multi-kernel neural network concurrent learning, monitoring, and forecasting system

  • US 6,647,377 B2
  • Filed: 04/09/2001
  • Issued: 11/11/2003
  • Est. Priority Date: 11/19/1997
  • Status: Expired due to Fees
First Claim
Patent Images

1. A method for determining whether refinement operations are indicated using learning correlations, comprising the steps of:

  • (a) receiving an iteration of measured input values for a current time trial;

    (b) providing a vector of input feature values based on the measured input values to a multi-kernel processor, each kernel of the processor operative for;

    receiving one or more of the input feature values, retrieving connection specifications, connection weights, and learning weights, and computing output feature values based on the received input feature values, the connection weights, and the connection specifications;

    (c) responding to a vector of computed output values based on the output feature values computed by each kernel; and

    (d) determining whether the refinement operations are indicated using the learning correlations and based on the output feature values.

View all claims
  • 5 Assignments
Timeline View
Assignment View
    ×
    ×