公开(公告)号：US11734570B1

公开(公告)日：2023-08-22

申请号：US16666850

申请日：2019-10-29

Applicant: Apple Inc.

Inventor： Daniel Kurz ,  Thomas Gebauer ,  Dewey H. Lee ,  Muhammad Ahmed Riaz ,  Qian Wang

IPC: G06N3/084 ,  G06F9/48 ,  G06N20/00 ,  G06N5/046

CPC classification number: G06N3/084 ,  G06F9/48 ,  G06N5/046 ,  G06N20/00

Abstract: The present disclosure describes techniques for training a neural network such that the trained network can be implemented to perform a utility task (e.g., a classification task) while inhibiting performance of a secondary task (e.g., a privacy-violating task). In some embodiments, the techniques include training a neural network using a first loss associated with a first task and a second loss associated with a second task different from the first task. In some embodiments, this includes performing a first training operation associated with the first loss, and performing a second training operation associated with the second loss, wherein the second training operation includes providing, to the neural network, a plurality of input items associated with the second task.

公开(公告)号：US11507836B1

公开(公告)日：2022-11-22

申请号：US17122327

申请日：2020-12-15

Applicant: Apple Inc.

Inventor： Daniel Kurz ,  Muhammad Ahmed Riaz

IPC: G06N3/08 ,  G06T7/50 ,  G06K9/62 ,  G06T7/70 ,  G10L15/24

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.