Adapting Bayesian network parameters on-line in a dynamic environment

• US 20030115325A1
• Filed: 12/18/2001
• Published: 06/19/2003
• Est. Priority Date: 12/18/2001
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Cohen, Ira, Bronstein, Alexandre, Duro, Marsha Prescott
• Granted Patent: US 7,685,278 B2
• Time in Patent Office: Days
• US Class Current: 709/225
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A method for adapting a Bayesian network includes determining a set of parameters for the Bayesian network, for example, initial parameters, and then updating the parameters in response to a set of observation data using an adaptive learning rate. The adaptive learning rate responds to any changes in the underlying modeled environment using minimal observation data.

Claims

1. A method for adapting a Bayesian network, comprising the steps of:

determining a set of parameters for the Bayesian network;

updating the parameters for the Bayesian network in response to a set of observation data using an adaptive learning rate.

2. The method of claim 1, wherein the step of updating the parameters comprises the steps of:

determining an initial value for the adaptive learning rate;

determining an estimate of the parameters in response to the observation data;

increasing the adaptive learning rate if an error between the estimate and a mean value of the parameters is relatively large.

3. The method of claim 1, wherein the step of updating the parameters comprises the steps of:

determining an initial value for the adaptive learning rate;

determining an estimate of the parameters in response to the observation data;

decreasing the learning rate when convergence is reached between the estimate and a mean value of the parameters.

4. The method of claim 1, further comprising the step of obtaining the observation data from an on-line environment.

5. The method of claim 1, wherein the step of obtaining comprises the step of obtaining a subset of values in the observation data from an on-line environment.

6. A system, comprising:

on-line environment that generates a set of observation data;

bayesian network that performs automated reasoning for the on-line environment in response to the observation data;

on-line adapter that adapts a set of parameters for the bayesian network in response to the observation data.

7. The system of claim 6, wherein the on-line adapter adapts the parameters by determining an initial set of the parameters and then updating the parameters in response to the observation data using an adaptive learning rate.

8. The system of claim 7, wherein the on-line adapter updates the parameters by determining an initial value for the adaptive learning rate and determining an estimate of the parameters in response to the observation data and then increasing the adaptive learning rate if an error between the estimate and a mean value of the parameters is relatively large.

9. The system of claim 7, wherein the on-line adapter updates the parameters by determining an initial value for the adaptive learning rate and determining an estimate of the parameters in response to the observation data and then decreasing the learning rate when convergence is reached between the estimate and a mean value of the parameters.

10. The system of claim 6, wherein the on-line adapter obtains a subset of values in the observation data from an on-line environment.