NONDETERMINISTIC TASK INITIATION BY A PERSONAL ASSISTANT MODULE

• US 20170351532A1
• Filed: 06/07/2016
• Published: 12/07/2017
• Est. Priority Date: 06/07/2016
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Li, Cheng, Wang, Bo, Kolak, Okan, Hodgson, Peter, Binay, Deniz, Amin, Dhruv, Gupta, Pravir, Shetti, Nitin, Rey, Javier
• Granted Patent: US 10,282,218 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 9/453 Help systems
• CPC Class Codes: G06F 9/4843 by program, e.g. task dispatcher, supervisor, operating system
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

Techniques are described herein for leveraging information about a user to enable a personal assistant module to make various inferences about what actions that may be responsive to a user declaration. In various implementations, upon identifying a user declaration received at a computing device, a plurality of candidate responsive actions that can be initiated by the computing device in response to the user declaration may be identified. A single candidate responsive action may then be non-deterministically (e.g., randomly, stochastically) selected to be exclusively initiated on the computing device in response to the user declaration.

Claims

1. A computer-implemented method, comprising:

identifying, by one or more processors, a user declaration received at a computing device;

identifying, by one or more of the processors, based on the user declaration, a plurality of candidate responsive actions that can be initiated by the computing device in response to the user declaration;

non-deterministically selecting, by one or more of the processors, a single candidate responsive action of the plurality of candidate responsive actions to be exclusively initiated on the computing device in response to the user declaration; and

exclusively initiating, by one or more of the processors, the single candidate responsive action on the computing device.

2. The computer-implemented method of claim 1, further comprising determining, by one or more of the processors, based on a history of interaction between the user and one or more computing devices operated by the user, a probability that initiation of each candidate responsive action will be responsive to the user declaration.

3. The computer-implemented method of claim 2, wherein the non-deterministically selecting is based at least in part on the probabilities of the plurality of candidate responsive actions.

4. The computer-implemented method of claim 3, further comprising altering, by one or more of the processors, a probability associated with the initiated responsive action based on one or more interactions between the user and the computing device following initiation of the selected action.

5. The computer-implemented method of claim 4, wherein the one or more interactions between the user and the computing device include rejection of the selected responsive action and manual initiation of an alternative candidate responsive action.

6. The computer-implemented method of claim 5, wherein the altering comprises altering the probability associated with the selected responsive action by an amount that is inversely proportionate to a measure of similarity between the selected responsive action and the alternative candidate responsive action.

7. The computer-implemented method of claim 2, wherein the probabilities are further based at least in part on a current context of the user.

8. The computer-implemented method of claim 7, wherein the current context of the user includes a current location of the user determined by the computing device.

9. The computer-implemented method of claim 7, wherein the current context of the user includes a current time at a current location of the user.

10. The computer-implemented method of claim 7, wherein a probability associated with at least one candidate responsive action is based on a measure of similarity between the user'"'"'s current context and a context of the user when the user initiated the at least one responsive action in the past.

11. The computer-implemented method of claim 1, wherein the plurality of candidate responsive actions include a plurality of software applications installed on the computing device.

12. The computer-implemented method of claim 1, wherein the user declaration comprises a request to initiate a telephone call, and the plurality of candidate responsive actions include a plurality of telephone applications installed on the computing device.

13. The computer-implemented method of claim 1, wherein the user declaration comprises a request to send a message, and the plurality of candidate responsive actions include a plurality of messaging applications installed on the computing device.

14. The computer-implemented method of claim 1, wherein the user declaration comprises a request to set a reminder, and the plurality of candidate responsive actions include a plurality of applications installed on the computing device that include reminder functionality.

15. The computer-implemented method of claim 1, wherein the user declaration comprises a request to initiate communication with a name, and the plurality of candidate responsive actions include a plurality of contacts of the user that include the name.

16. A system comprising one or more processors and memory storing instructions that cause the one or more processors to:

identify a user declaration received at a computing device;

identify, based on the user declaration, a plurality of candidate responsive actions that can be initiated by the computing device to potentially satisfy the user declaration;

non-deterministically select a single candidate responsive action of the plurality of candidate responsive actions to be exclusively initiated on the computing device in response to the user declaration; and

exclusively initiate the single candidate responsive action on the computing device.

17. The system of claim 16, further comprising instructions to determine, based on a history of interaction between the user and one or more computing devices operated by the user, a probability that initiation of each candidate responsive action will satisfy the user declaration;

wherein the non-deterministically selection is based on the plurality of probabilities.

18. The system of claim 17, further comprising instructions to alter the probability associated with the-single candidate responsive action based on one or more interactions between the user and the computing device following initiation of the single candidate responsive action.

19. The system of claim 18, wherein the one or more interactions between the user and the computing device include rejection of the selected responsive action and manual initiation of an alternative candidate responsive action.

20. At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by a computing system, cause the computing system to perform the following operations:

identifying a user declaration received at a computing device;

identifying, based on the user declaration, a plurality of candidate responsive actions that can be initiated by the computing device in response to the user declaration;

non-deterministically selecting a single candidate responsive action of the plurality of candidate responsive actions to be exclusively initiated on the computing device in response to the user declaration; and

exclusively initiating the single candidate responsive action on the computing device.