Federated learning using local ground truth estimation
AssigneeApple Inc.
InventorsDaniel Kurz, Muhammad Ahmed Riaz
Life of this patent
Estimating the term…
Abstract
Various implementations disclosed herein include devices, systems, and methods that involve federated learning techniques that utilize locally-determined ground truth data that may be used in addition to, or in the alternative to, user-provided ground truth data. Some implementations provide an improved federated learning technique that creates ground truth data on the user device using a second prediction technique that differs from a first prediction technique/model that is being trained. The second prediction technique may be better but may be less suited for real time, general use than the first prediction technique.
Claims25
A method comprising:
- at an electronic device having a processor:
- generating a first prediction using a first prediction technique and locally-obtained data, the first prediction technique using a first machine learning model, the first machine learning model distributed by a second device to a plurality of user devices;
- generating a second prediction using a second prediction technique and the locally-obtained data;
- determining a machine learning model update for the first machine learning model using the second prediction technique to provide ground truth, wherein the machine learning model update is determined based on the first prediction and the second prediction;
- and
- providing the machine learning model update to the second device, wherein the second device is configured to create an updated machine learning model based on the machine learning model update and distribute the updated machine learning model to the plurality of user devices.
A device comprising:
- a non-transitory computer-readable storage medium;
- and
- one or more processors coupled to the non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium comprises program instructions that, when executed on the one or more processors, cause the system to perform operations comprising:
- generating a first prediction using a first prediction technique and locally-obtained data, the first prediction technique using a first machine learning model, the first machine learning model distributed by a second device to a plurality of user devices;
- generating a second prediction using a second prediction technique and the locally-obtained data;
- determining a machine learning model update for the first machine learning model using the second prediction technique to provide ground truth, wherein the machine learning model update is determined based on the first prediction and the second prediction;
- and
- providing the machine learning model update to the second device, wherein the second device is configured to create an updated machine learning model based on the machine learning model update and distribute the updated machine learning model to the plurality of user devices.
A non-transitory computer-readable storage medium, storing program instructions computer-executable on a computer to perform operations comprising:
- generating a first prediction using a first prediction technique and locally-obtained data, the first prediction technique using a first machine learning model, the first machine learning model distributed by a second device to a plurality of user devices;
- generating a second prediction using a second prediction technique and the locally-obtained data;
- determining a machine learning model update for the first machine learning model using the second prediction technique to provide ground truth, wherein the machine learning model update is determined based on the first prediction and the second prediction;
- and
- providing the machine learning model update to the second device, wherein the second device is configured to create an updated machine learning model based on the machine learning model update and distribute the updated machine learning model to the plurality of user devices.
Dependencies are inferred from the claim text.