Data processing apparatus, data processing method, and computer program for setting targets in a range and taking action to attain the target

• US 8,195,586 B2
• Filed: 03/27/2009
• Issued: 06/05/2012
• Est. Priority Date: 05/28/2008
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Aoyama, Kazumi, Ito, Masato, Noda, Kuniaki
• Primary Examiner(s): CHEN, ALAN S
• Application Number:
  US12/413,004
• Publication Number:
  US 20090299933A1
• Time in Patent Office: 1,166 Days
• Field of Search: None
• US Class Current: 706/15
• CPC Class Codes: G06N 3/049 Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs

Abstract

A data processing apparatus includes: predicting means for calculating a prediction value of time series data with respect to input of the time series data using a prediction model for predicting the time series data; determining means for determining a target value of the time series data on the basis of the prediction value of the time series data; error calculating means for calculating an error of the prediction value relative to the target value; and retrieving means for retrieving error reduction data as input of the time series data to the prediction model for reducing the error of the prediction value.

Claims

1. A data processing apparatus comprising:

a predicting unit configured to calculate a prediction value of time series data with respect to input of the time series data using a prediction model for predicting the time series data;

a determining unit configured to determine a target value of the time series data using the prediction value of the time series data;

an error calculating unit configured to calculate, via a processor, an error of the prediction value relative to the target value; and

a retrieving unit configured to retrieve and input error reduction data as part of the time series data input into the prediction model to reduce the error of the prediction value.

2. The data processing apparatus according to claim 1, wherein the determining unit determines, as the target value, a function value of a function that takes as an argument the prediction value.

3. The data processing apparatus according to claim 2, further comprising:

an updating unit configured to update a parameter of the prediction model as learning of the prediction model using the error reduction data.

4. The data processing apparatus according to claim 2, wherein

the time series data includes action data that causes an agent to act and sensor data representing a state of an environment,

the determining unit determines a target value of the sensor data, and

the retrieving unit retrieves, as the error reduction data, the time series data including the action data for reducing an error of the prediction value of the sensor data relative to the target value.

5. The data processing apparatus according to claim 4, wherein the sensor data the state of the environment includes at least one of a photographed image, environment sound, or a distance of the agent from an object.

6. The data processing apparatus according to claim 4, wherein the sensor data further includes a state of the agent.

7. The data processing apparatus according to claim 6, wherein the state of the agent includes at least one of a rotation angle or angular velocity of an actuator of the agent.

8. The data processing apparatus according to claim 4, wherein the target value is determined by using the prediction value of only the sensor data or only the action data.

9. The data processing apparatus according to claim 8, wherein the determining unit determines the target value from a portion of the sensor data or the action data.

10. The data processing apparatus according to claim 2, wherein the prediction model is a function approximator that performs function approximation to approximate a function.

11. The data processing apparatus according to claim 2, wherein the function includes multiplying the predication value by a predetermined multiplier to obtain the target value.

12. The data processing apparatus according to claim 1, wherein the prediction model is a Support Vector Regression.

13. A data processing apparatus comprising:

predicting means for calculating a prediction value of time series data with respect to input of the time series data using a prediction model for predicting the time series data;

determining means for determining a target value of the time series data using the prediction value of the time series data;

error calculating means for calculating an error of the prediction value relative to the target value; and

retrieving means for retrieving and inputting error reduction data as part of the time series data input into the prediction model to reduce the error of the prediction value.

14. A data processing method comprising the steps of:

calculating a prediction value of time series data with respect to input of the time series data using a prediction model for predicting the time series data;

determining a target value of the time series data using the prediction value of the time series data;

calculating, via a processor, an error of the prediction value relative to the target value; and

retrieving and inputting error reduction data as part of the time series data input into the prediction model to reduce the error of the prediction value.

15. A non-transitory computer-readable medium storing computer readable instructions thereon that when executed by a data processing apparatus cause the data processing apparatus to perform a method comprising:

calculating a prediction value of time series data with respect to input of the time series data using a prediction model for predicting the time series data;

determining a target value of the time series data using the prediction value of the time series data;

for calculating an error of the prediction value relative to the target value; and

retrieving and inputting error reduction data as part of the time series data input into the prediction model to reduce the error of the prediction value.