Semi-autonomous intelligent task hub

Patent metadata

Field	Value
Publication number	US-11416290-B2
Application number	US-202016886764-A
Country	US
Kind code	B2
Filing date	May 28, 2020
Priority date	May 28, 2020
Publication date	Aug 16, 2022
Grant date	Aug 16, 2022

Abstract

The present disclosure relates to systems and methods for an interactive, intelligent hub built around the completion of a task. This hub brings together resources, information, suggested steps, and other automated assistance to facilitate the completion of the task. AI-based assistance may indicate which steps can be completed by automated processes, and dispatch those processes, or suggest resources to assist in the completion of other steps. The hub displays the current status of the task, and lives until the completion of the task, or abandonment by the user.

First claim

What is claimed is:

1. A method of semi-autonomously managing a task, the method comprising:
determining that the task comprises a plurality of subtasks including a first subtask and a second subtask;
determining by a machine learning model that the first subtask is automatable based on a state of the plurality of subtasks and a definition for the first subtask;
automatically performing the first subtask based on the determination that the first subtask is automatable;
determining by the machine learning model that the second subtask requires user input based on the state of the plurality of subtasks and a definition for the second subtask; and
notifying a user that the user input is needed to complete the second subtask.

2. The method of claim 1 further comprising:
determining that the second subtask is not automatable based on an empty slot in the definition for the second subtask.

3. The method of claim 1 further comprising:
determining that the first subtask is automatable based on satisfaction of a dependency between the first subtask and one or more of the plurality of subtasks.

4. The method of claim 1 further comprising:
determining by the machine learning model an order for performing the plurality of subtasks.

5. The method of claim 1 wherein the plurality of subtasks further comprises a third subtask and the method further comprises:
determining by the machine learning model that the third subtask should be delegated to a third party based on the state of the plurality of subtasks and a definition of the third subtask; and
notifying the third party that the third subtask has been delegated to the third party.

6. The method of claim 5 further comprising:
receiving confirmation from the third party that the third subtask is complete; and
updating a state of the third subtask.

7. The method of claim 1 wherein the machine learning model is a neural network trained using task completion data.

8. The method of claim 1 wherein the machine learning model is one of a recurrent neural network, a transformer network, a multi-task neural network, and a bidirectional recurrent neural network.

9. The method of claim 1 further comprising:
receiving the user input from the user for the second subtask;
updating a state of the second subtask; and
updating the definition of the second subtask.

10. The method of claim 9 further comprising:
determining by the neural network that the second subtask is automatable based the updated state of the second subtask and the updated definition for the second subtask; and
automatically performing the second subtask based on the determination that the second subtask is automatable.

11. A system for semi-autonomously managing a task comprising:
a processor;
memory storing computer executable instructions that when executed cause the processor to:
determine that the task comprises a plurality of subtasks including a first subtask and a second subtask;
determine by a machine learning model that the first subtask is automatable based on a state of the plurality of subtasks and a definition for the first subtask;
automatically perform the first subtask based on the determination that the first subtask is automatable;
determine by the machine learning model that the second subtask requires user input based on the state of the plurality of subtasks and a definition for the second subtask; and
notify a user that the user input is needed to complete the second subtask.

12. The system of claim 11 wherein the machine learning model is one of a recurrent neural network, a transformer network, a multi-task neural network, and a bidirectional recurrent neural network.

13. The system of claim 12 further comprising:
determining an order to complete the plurality of subtasks wherein the second subtask is scheduled to be completed before the first subtask.

14. The system of claim 11 further comprising computer executable instructions that when executed cause the processor to:
determine that the first subtask is automatable based on an empty slot in the definition for the first subtask.

15. The system of claim 11 further comprising computer executable instructions that when executed cause the processor to:
determine that the first subtask is automatable based on satisfaction of a dependency between the first subtask and one or more of the plurality of subtasks.

16. A method of semi-autonomously managing a task comprising a plurality of subtasks, the method comprising:
determining by a neural network that a first subtask of the plurality of subtasks is automatable based on inputs to the neural network, wherein the inputs comprise a state of the plurality of subtasks and definitions for plurality of subtasks;
automatically performing the first subtask based on the determination that the first subtask is automatable;
determining by the neural network that a second subtask requires user input based the inputs; and
notifying a user that the user input is needed to complete the second subtask.

17. The method of claim 16 further comprising:
determining by the neural network an order of the plurality of subtasks based on the inputs.

18. The method of claim 16 wherein the inputs further comprise a dependency between the plurality of subtasks.

19. The method of claim 16 further comprising delegating by the neural network a third task to a third party based on the inputs.

20. The method of claim 16 further comprising identifying by the neural network a resource needed to complete the second subtask.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Sim Robert Alexander
• White Ryen William
• Shaya Omar
• Plontsch Bernd Ingo
• Nouri Elnaz

Classifications

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/0455

  Auto-encoder networks; Encoder-decoder networks · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/09

  Supervised learning · CPC title