Predictive controller for applications

• US 10,552,752 B2
• Filed: 11/02/2015
• Issued: 02/04/2020
• Est. Priority Date: 11/02/2015
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Kashyap, Abhinav
• Primary Examiner(s): Afshar, Kamran
• Assistant Examiner(s): Buss, Benjamin J
• Application Number:
  US14/930,563
• Publication Number:
  US 20170124474A1
• Time in Patent Office: 1,555 Days
• Field of Search: None
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A “Predictive Controller” operates with any type of controller or user input device to predict user inputs or responses to a current state of an application. A predictive model of the current state of the application is applied to prior user inputs to jointly predict a current user-specific psychological state or profile of the user and a predicted next user response or input. The predicted response or input is provided as the user input to the particular application prior to receiving the actual user input, thereby reducing latency of the response of the application to that actual user input. In addition, a tangible feedback corresponding to the predicted next user input is provided. Further, the predictive capabilities of the Predictive Controller can be applied to locally or remotely hosted instances of the application to reduce latencies associated with user inputs received from any type of controller or user input device.

Claims

1. A system, comprising:

a hardware controller device that receives user inputs for an application during execution of the application;

a trained predictive model corresponding to each of a plurality of possible execution states of the application;

an application module that determines a current execution state of the application;

an input prediction module that applies the predictive model corresponding to the current execution state of the application to one or more prior user inputs to jointly predict a current psychological state of the user and a corresponding predicted next user input; and

an application module that receives the next predicted user input from the input prediction module and that applies the predicted next user input to the application.

2. The system of claim 1 further comprising a model update module that:

compares the predicted next user input to a corresponding actual user input received from the hardware controller device subsequent to applying the predicted next user input to the application; and

if the predicted next user input does not match the actual user input, updates the corresponding predictive model based on the actual user input.

3. The system of claim 1 further comprising a user matching module that matches a plurality of users for participation in a shared instance of the application based on the predicted psychological state of each user.

4. The system of claim 1 further comprising a model training module that:

applies a machine-learning process to train each predictive model from anonymized user inputs of a plurality of users received during each possible execution state of the application; and

categorizes the anonymized user inputs into a plurality of possible psychological states for each corresponding execution state of the application.

5. The system of claim 1 wherein the trained predictive model corresponding to each of the plurality of possible execution states of the application further comprises a Hidden Markov Model (HMM) adapted to jointly transition between a prediction of the possible psychological states and to predict the corresponding next user input as an ongoing process throughput execution of the application.

6. The system of claim 1 further comprising an application adaptation module that automatically adapts application execution to influence the predicted current psychological state of the user.

7. The system of claim 1 further comprising a feedback module that initiates a tangible notification corresponding to the predicted next user input.

8. The system of claim 7, wherein the tangible notification further comprises causing the hardware controller to generate a physically tactile response in a sequence of one or more controller buttons representing the predicted next user input.

9. A computer-implemented process, comprising:

receiving one or more prior user inputs to an application via a hardware controller device;

for a current application state, applying a corresponding predictive state-based application model to the prior user inputs to jointly predict a current user psychological state and a next user input;

applying the predicted next user input to the application prior to receipt of an actual next user input; and

following receipt of the actual next user input, initiating a tangible feedback to indicate any difference between predicted next user input and actual next user input.

10. The computer-implemented process of claim 9 further comprising comparing the predicted next user input to actual next user input and, in the event that the prediction does not match the actual input, updating the corresponding predictive state-based application model.

11. The computer-implemented process of claim 9 further comprising:

receiving one or more anonymized sets of hardware controller data from a plurality of users for each of a plurality of application states;

for each application state, clustering the corresponding anonymized hardware controller data into sets of psychological behavior states; and

for each application state, applying a machine-learning process to generate the corresponding predictive state-based application model.

12. The computer-implemented process of claim 9 further comprising matching a plurality of users for participation in a shared instance of the application based on the predicted psychological state of each user.

13. The computer-implemented process of claim 9 wherein the predictive state-based application model corresponding to the current application state further comprises a Hidden Markov Model (HMM) adapted to jointly transition between a prediction of possible user psychological states and to predict the corresponding next user input as an ongoing process throughput execution of the application.

14. The computer-implemented process of claim 9 further comprising automatically adapting application execution to influence the predicted current user psychological state.

15. The computer-implemented process of claim 9 further comprising rendering a visible alert representing the predicted next user input on one or more display devices on which the application is being rendered.

16. A computer-readable storage device having computer executable instructions stored therein, said instructions causing a computing device to execute a method comprising:

executing an application;

receiving one or more prior user inputs relating to a current application state of the application from a hardware controller device;

applying a predictive model corresponding to the current application state to the prior user inputs to jointly predict a current psychological state of the user and a corresponding predicted next user input; and

applying the predicted next user input to the application prior to receiving a corresponding actual user input.

17. The computer-readable storage device of claim 16 further comprising:

comparing the predicted next user input to the corresponding actual user input; and

if the predicted next user input does not match the actual user input, updating the corresponding predictive model based on the actual user input.

18. The computer-readable storage device of claim 16 further comprising matching a plurality of users for participation in a shared instance of the application based on the predicted psychological state of each user.

19. The computer-readable storage device of claim 16 further comprising automatically adapting application execution to influence the predicted current psychological state of the user.

20. The computer-readable storage device of claim 16 wherein the trained predictive model comprises a Hidden Markov Model (HMM) adapted to jointly transition between a prediction of possible psychological states and to predict the corresponding next user input as an ongoing process throughput execution of the application.