System, method, and program storage medium for executing a learning algorithm

• US 6,263,325 B1
• Filed: 09/29/1998
• Issued: 07/17/2001
• Est. Priority Date: 02/27/1998
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Yoshida, Yukiko, Adachi, Nobue
• Primary Examiner(s): Powell, Mark R.
• Assistant Examiner(s): Khatri, Anil
• Application Number:
  US09/161,984
• Time in Patent Office: 1,022 Days
• Field of Search: 706/12, 706/13, 706/14, 706/16, 706/25
• US Class Current: 706/12
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

A system for executing a learning algorithm for solving an optimization problem of an evolutionary algorithm, etc. In this system, a learning algorithm executing device and a terminal device are interconnected by a network. In the terminal device, a change of an execution process of the learning algorithm may be set during the execution of the learning algorithm. In the learning algorithm executing device, the execution of the learning algorithm is continued according to the change of the execution process, which is set in the terminal device. As the change of the execution process of the learning algorithm, an execution condition of the learning algorithm is changed during its execution, or a visual or numerical change is made to an object representing the progress state of the execution of the algorithm on a display screen, in the terminal device.

Claims

1. A system for executing a learning algorithm for solving an optimization problem, wherein the system improves performance of the learning algorithm, comprising:

a change processing device setting a change of an execution process of the learning algorithm during execution of the learning algorithm; and

an executing device continuing the execution of the learning algorithm according to the change of the execution process, if the change of the execution process is set during the execution of the learning algorithm.

2. The system according to claim 1, wherein:

said change processing device sets a new execution condition of the learning algorithm during the execution of the learning algorithm; and

said executing device continues the execution of the learning algorithm under the new execution condition.

3. The system according to claim 1, wherein:

said change processing device sets a new step progress controlling method of the learning algorithm during the execution of the learning algorithm; and

said executing device continues the execution of the learning algorithm with the new step progress controlling method.

4. The system according to claim 1, further comprising:

a display device visualizing and displaying a progress state of the execution of the learning algorithm by using an object, wherein;

said change processing device makes a visual or numerical change to the object displayed by said displaying means on a display screen; and

said executing device executes the learning algorithm, after contents of the change made to the object are reflected in a current learning algorithm, if the change is made to the object.

5. The system according to claim 1, wherein:

said executing device is included in both a terminal device and a learning algorithm executing device; and

the terminal device and the learning algorithm executing device are interconnected by a network.

6. The system according to claim 5, wherein:

the learning algorithm executing device further comprises a transmitting unit for transmitting to the terminal device a progress state of the execution process of the learning algorithm, which is performed by said executing device; and

the terminal device further comprises a transmitting unit for transmitting to the learning algorithm executing device the change of the execution process of the learning algorithm, which is set by said change processing device.

7. The system according to claim 1, wherein:

the learning algorithm is an evoluntionary algorithm for obtaining an optimum solution or promising solutions to a given problem by evolving solution candidates.

8. A device for controlling execution of a learning algorithm for solving an optimization problem, wherein the device improves performance of the learning algorithm, comprising:

a change processing unit setting a change of an execution process of the learning algorithm during the execution of the learning algorithm; and

a transmitting unit transmitting to a learning algorithm executing device information about the change of the execution process of the learning algorithm, which is set by said change processing unit.

9. The device according to claim 8, further comprising:

a displaying unit visualizing and displaying a progress state of the execution of the learning algorithm, which is received from the learning algorithm executing device.

10. A device for executing a learning algorithm for solving an optimization problem, wherein the device improves performance of the learning algorithm, comprising:

a receiving unit receiving information about a change of an execution process of the learning algorithm if the change is made to the execution process during the execution of the learning algorithm; and

an executing unit continuing the execution of the learning algorithm according to the information about the change of the execution process, which is received by said receiving unit.

11. The device according to claim 10, further comprising:

an outputting unit outputting to a terminal a progress state of the execution process of the learning algorithm, which is performed by said executing unit.

12. A device for controlling execution of a learning algorithm for solving an optimization problem, wherein the device improves performance of the learning algorithm comprising:

a processing unit setting at least either of an execution condition and a step progress controlling method of the learning algorithm; and

a transmitting unit transmitting to a learning algorithm executing device at least either of the execution condition and the step progress controlling method, which is set by said processing unit.

13. The device according to claim 12, further comprising:

a displaying unit visualizing and displaying a progress state of the execution of the learning algorithm, which is received from the learning algorithm executing device.

14. A device for executing a learning algorithm for solving an optimization problem, wherein the device improves performance of the learning algorithm, comprising:

a receiving unit receiving from a terminal device at least either information of an execution condition and a step progress controlling method of the learning algorithm; and

an executing unit continuing the execution of the learning algorithm according to at least either information of the execution condition and the step progress controlling method of the learning algorithm, which is received by said receiving unit.

15. The device according to claim 14, further comprising:

an outputting unit outputting to the terminal the progress state of an execution process of the learning algorithm, which is performed by said executing unit.

16. A method for executing a learning algorithm for solving an optimization problem, wherein the method improves performance of the learning algorithm, comprising:

setting a change of an execution process of the learning algorithm during the execution of the learning algorithm; and

continuing the execution of the learning algorithm according to the change of the execution process, if the change of the execution process is set during the execution of the learning algorithm.

17. The method according to claim 16, wherein:

a new execution condition of the learning algorithm is set during the execution of the learning algorithm in said setting; and

the execution of the learning algorithm is continued under the new execution condition in said continuing.

18. The method according to claim 16, wherein:

a new step progress controlling method of the learning algorithm is set during the execution of the learning algorithm in said setting; and

the execution of the learning algorithm is continued with the new step progress controlling method in said continuing.

19. The method according to claim 16, further comprising visualizing and displaying a progress state of the execution of the learning algorithm, wherein:

a visual or numerical change is made to a displayed object in said setting; and

the learning algorithm is executed after contents of the change made to the object are reflected in a current learning algorithm in said continuing.

20. A method for controlling execution of a learning algorithm for solving an optimization problem, wherein the method improves performance of the learning algorithm, comprising:

setting a change of an execution process of the learning algorithm during the execution of the learning algorithm; and

transmitting to a learning algorithm executing device information about the set change of the execution process of the learning algorithm.

21. The method according to claim 20, further comprising visualizing and displaying a progress state of the execution of the learning algorithm, which is received from the learning algorithm executing device.

22. A method for executing a learning algorithm for solving an optimization problem, wherein the method improves performance of the learning algorithm, comprising:

receiving information about a change of an execution process of the learning algorithm from a terminal device, if the change is made to the execution process during the execution of the learning algorithm; and

continuing the execution of the learning algorithm according to the received information about the change of the execution process.

23. The method according to claim 22, further comprising outputting to the terminal device a progress state of the execution process of the learning algorithm.

24. A method for controlling execution of a learning algorithm for solving an optimization problem, wherein the method improves performance of the learning algorithm, comprising:

setting at least either of an execution condition and a step progress controlling method of the learning algorithm; and

transmitting to a learning algorithm executing device at least either information of the execution condition information and the step progress controlling method of the learning algorithm, which is set in the setting, the transmitting performed while the learning algorithm is executing.

25. The method according to claim 24, further comprising visualizing and displaying a progress state of the execution of the learning algorithm, which is received from the learning algorithm executing device.

26. A method for executing a learning algorithm for solving an optimization problem, wherein the method improves performance of the learning algorithm, comprising:

receiving from a terminal device at least either information of an execution condition and a step progress controlling method of the learning algorithm; and

continuing the execution of the learning algorithm according to at least either information of the execution condition and the step progress controlling method of the learning algorithm, which is received from the terminal device.

27. The method according to claim 26, further comprising outputting to the terminal device a progress state of an execution process of the learning algorithm.

28. A computer-readable medium storing a program implementing a method for executing a learning algorithm for solving an optimization problem by using a computer, wherein the method improves performance of the learning algorithm, the method comprising:

setting a change of an execution process of the learning algorithm during execution of the learning algorithm; and

continuing the execution of the learning algorithm according to the change of the execution process, if the change of the execution process is set during the execution of the learning algorithm.

29. The medium according to claim 28, wherein:

a new execution condition of the learning algorithm is set during the execution of the learning algorithm in said setting; and

the execution of the learning algorithm is continued under the new execution condition in said continuing.

30. The medium according to claim 28, wherein:

a new step progress controlling method of the learning algorithm is set during the execution of the learning algorithm in said setting; and

the execution of the learning algorithm is continued with the new step progress controlling method in said continuing.

31. The medium according to claim 28, further comprising visualizing and displaying a progress state of the execution of the learning algorithm by using an object, wherein

a visual or numerical change is made to a displayed object in said setting,

the learning algorithm is executed after contents of the change made to the object are reflected in a current learning algorithm in said continuing.

32. A computer-readable medium storing a program implementing a method for executing a learning algorithm solving an optimization problem by using a computer, wherein the method improves performance of the learning algorithm, the method comprising:

setting a change of an execution process of the learning algorithm during execution of the learning algorithm; and

transmitting to a learning algorithm executing device information about the set change of the execution process of the learning algorithm.

33. The medium according to claim 32, further comprising a visualizing and displaying a progress state of the execution of the learning algorithm, which is received from the learning algorithm executing device.

34. A computer-readable medium storing a program for implementing a method for executing a learning algorithm for solving an optimization problem by using a computer, wherein the method improves performance of the learning algorithm, the method comprising:

receiving from a terminal device information about a change of an execution process, if the change is made to the execution process of the learning algorithm during the execution of the learning algorithm; and

continuing the execution of the learning algorithm according to the received information about the change of the execution process.

35. The medium according to claim 34, further comprising outputting to the terminal device a progress state of the execution process of the learning algorithm.

36. A computer-readable medium storing a program for implementing a method for executing a learning algorithm for solving an optimization problem by using a computer, wherein the method improves performance of the learning algorithm, the method comprising:

setting at least either of an execution condition and a step progress controlling method of the learning algorithm; and

transmitting to a learning algorithm executing device at least either information of the execution condition and the step progress controlling method of the learning algorithm, which is set, the transmitting performed while the learning algorithm is executing.

37. The medium according to claim 36, further comprising visualizing and displaying a progress state of the execution of the learning algorithm, which is received from the learning algorithm executing device in the setting.

38. A computer-readable medium storing a program for implementing a method for executing a learning algorithm for solving an optimization problem by using a computer, wherein the method improves performance of the learning algorithm, the method comprising:

receiving from a terminal device at least either information of an execution condition and a step progress controlling method of the learning algorithm; and

continuing the execution of the learning algorithm according to either information of the execution condition and the step progress controlling method of the learning algorithm, which is received from the terminal device.

39. The medium according to claim 38, further comprising outputting to the terminal device a progress state of an execution process of the learning algorithm.