Edge devices utilizing personalized machine learning and methods of operating the same

At least one non-transitory machine readable medium comprising instructions that, when executed, cause at least one processor to at least:

• access a first machine learning model from a cloud service provider;
• collect local user data;
• train the first machine learning model to create a second machine learning model using the local user data;
• access permissions indicating constraints on the local user data concerning access to the local user data outside of an edge device;
• apply the permissions to the local user data to create a sub-set of the local user data to be shared outside of the edge device;
• determine whether to share the second machine learning model based on the permissions and metadata associated with the local user data used to create the second machine learning model;
• provide the sub-set of the local user data to a public data repository;
• and provide the second machine learning model to the public data repository in response to determining that the second machine learning model is to be shared.

An edge device comprising:

• means for accessing a first machine learning model from a cloud service provider;
• means for collecting local user data;
• means for training the first machine learning model to create a second machine learning model using the local user data;
• means for accessing permissions indicating constraints on the local user data with respect to sharing outside of the edge device;
• means for applying the permissions to the local user data to create a sub-set of the local user data to be shared outside of the edge device, the means for applying is further to determine whether to share the second machine learning model based on the permissions and metadata based on the local user data used to create the second machine learning model;
• and
• means for providing the sub-set of the local user data to a public data repository, the means for providing is further to, in response to determining that the second machine learning model is to be shared, provide the second machine learning model to the public data repository.

A method of using a personalized machine learning model, the method comprising:

• accessing a first machine learning model from a cloud service provider;
• collecting, using a processor of an edge device, local user data;
• training the first machine learning model to create a second machine learning model using the local user data;
• accessing permissions indicating constraints on the local user data concerning access to the local user data outside of the edge device;
• applying the permissions to the local user data to create a sub-set of the local user data to be shared outside of the edge device;
• determining whether to share the second machine learning model based on the permissions and metadata associated with the local user data used to create the second machine learning model;
• providing the sub-set of the local user data to a public data repository;
• and
• providing the second machine learning model to the public data repository in response to determining that the second machine learning model is to be shared.