Systems, devices, and methods for machine learning using a distributed framework

1. A method of machine learning using a distributed framework, the method comprising:

selecting, by a computing device, at least a remote device of a plurality of remote devices;

determining, by the computing device, a confidence level of the at least a remote device; and

assigning, by the computing device, at least a machine-learning task to the at least a remote device, wherein assigning further comprises:

assigning at least a secure data storage task to the at least a remote device; and

assigning at least a model-generation task to the at least a remote device, wherein assigning the at least a model-generation task further comprises assigning the at least a model generation task to at least a device of the at least a remote device that performs attested computing.

2. The method of claim 1 , wherein selecting the at least a remote device further comprises selecting a plurality of proximate devices.

3. The method of claim 1 , wherein assigning the at least a model-generation task further comprises assigning a first machine-learning task to a first device of the at least a remote device and assigning a second machine-learning task to a second device of the at least a remote device.

4. The method of claim 1 , further comprising evaluating the attested computing of the at least a device.

5. The method of claim 1 further comprising:

dividing data to be stored in the at least a data storage task into a first data lot and a second data lot;

assigning the first data lot to a first data storage device of the at least a remote device; and

assigning the second data lot to a second data storage device of the at least a remote device.

6. The method of claim 5 , wherein:

the at least a machine learning task further comprises at least a first model-generation task and at least a second model-generation task; and

the first data lot contains data for the first model-generation task and the second data lot contains data for the second model-generation task.

7. The method of claim 6 further comprising assigning the first model-generation task to a first model-generation device that is proximate to the first data storage device and assigning the second model-generation task to a second model-generation device that is proximate to the second data storage device.

8. The method of claim 1 further comprising encrypting data to be stored in the at least a data storage task.

9. The method of claim 1 , wherein:

the at least a model-generation task is assigned to at least a model-generation device of the at least a remote device;

the at least a data storage task is assigned to at least a data storage device of the at least a remote device; and

the at least a model-generation device has a higher confidence level than the at least a data storage device.

10. A system for machine learning using a distributed framework, the system comprising a computing device communicatively connected to a plurality of remote devices, the computing device designed and configured to:

select at least a remote device of a plurality of remote devices, determine a confidence level of the at least a remote device, and assign at least a machine-learning task to the at least a remote device, wherein assigning further comprises assigning at least a secure data storage task to the at least a remote device and assigning at least a model-generation task to the at least a remote device, wherein assigning the at least a model-generation task further comprises assigning the at least a model generation task to at least a device of the at least a remote device that performs attested computing.

11. The system of claim 10 , wherein the computing device is further configured to select the at least a remote device by selecting a plurality of proximate devices.

12. The system of claim 10 , wherein assigning the at least a model-generation task further comprises assigning a first machine-learning task to a first device of the at least a remote device and assigning a second machine-learning task to a second device of the at least a remote device.

13. The system of claim 10 , wherein the computing device is further configured to evaluate the attested computing of the at least a device.

14. The system of claim 10 , wherein the computing device is further configured to:

divide data to be stored in the at least a data storage task into a first data lot and a second data lot;

assign the first data lot to a first data storage device of the at least a remote device; and

assign the second data lot to a second data storage device of the at least a remote device.

15. The system of claim 14 , wherein:

the at least a machine learning task further comprises at least a first model-generation task and at least a second model-generation task; and

the first data lot contains data for the first model-generation task and the second data lot contains data for the second model-generation task.

16. The system of claim 15 wherein the computing device is further configured to assign the first model-generation task to a first model-generation device that is proximate to the first data storage device and assigning the second model-generation task to a second model-generation device that is proximate to the second data storage device.

17. The system of claim 10 wherein the computing device is further configured to encrypt data to be stored in the at least a data storage task.

18. The system of claim 10 , wherein:

the at least a model-generation task is assigned to at least a model-generation device of the at least a remote device;

the at least a data storage task is assigned to at least a data storage device of the at least a remote device; and

the at least a model-generation device has a higher confidence level than the at least a data storage device.