公开(公告)号：US10915663B1

公开(公告)日：2021-02-09

申请号：US16261112

申请日：2019-01-29

Applicant: Facebook, Inc.

Inventor： Cristian Canton Ferrer ,  Brian Dolhansky ,  Phong Dinh ,  Bryan Wu ,  Zhen Ling Tsai ,  Eric Erkon Hsin

IPC: G06K9/00 ,  G06F21/62 ,  G06N3/08 ,  G06K9/62

Abstract: Systems, methods, and non-transitory computer-readable media can be configured to train a featurizer based at least in part on a set of training data. The featurizer can be applied to at least one input to generate at least one tensor. The at least one tensor obfuscates or excludes at least one feature in the at least one input.

公开(公告)号：US20210141926A1

公开(公告)日：2021-05-13

申请号：US16790437

申请日：2020-02-13

Applicant: Facebook, Inc.

Inventor： Cristian Canton Ferrer ,  Brian Dolhansky ,  Hao Guo ,  Eric Erkon Hsin ,  Phong Dinh

IPC: G06F21/62 ,  G06K9/62 ,  G06K9/00 ,  G06N3/08 ,  G06N20/20

Abstract: In one embodiment, a method includes accessing a first machine-learning model trained to generate a feature representation of an input data, a second machine-learning model trained to generate a desired result based on the feature representation, and a third machine-learning model trained to generate an undesired result based on the feature representation, and training a fourth machine-learning model by generating a secured feature representation by processing a first output of the first machine-learning model using the fourth machine-learning model, generating a second output and a third output by processing the secured feature representation using, respectively, the second and third machine-learning models, and updating the fourth machine-learning model according to an optimization function configured to optimize a correctness of the second output and an incorrectness of the third output.