公开(公告)号：US20240054225A1

公开(公告)日：2024-02-15

申请号：US17316296

申请日：2021-05-10

Applicant: Facebook, Inc.

Inventor： Luhui Hu ,  Can Lin ,  Xiaohua Jiang

IPC: G06F21/57 ,  G06F11/36 ,  G06F8/77

CPC classification number: G06F21/577 ,  G06F11/3688 ,  G06F8/77 ,  G06F2221/033

Abstract: A method and system are described. The method includes determining, in a development phase of a software service, whether the software service complies with a first policy in response to a request. The method also monitors, in at least one of a testing phase or a production phase of the software service, whether operation of the software service complies with a second policy. Based on the determining and the monitoring, an indication of a service vulnerability is generated in response to the software service failing to comply with the first policy in the development phase or failing to comply with the second policy in the at least one of the testing phase or the production phase.

公开(公告)号：US20210406779A1

公开(公告)日：2021-12-30

申请号：US16913054

申请日：2020-06-26

Applicant: Facebook, Inc.

Inventor： Luhui Hu ,  Yinglong Xia

IPC: G06N20/20 ,  G06N5/02 ,  G06N5/04 ,  G06N3/02

Abstract: In one embodiment, a computing system may receive query information associated with a machine-learning model. The system may access a knowledge graph that defines relationships between a number of machine-learning models and a number of features of the machine-learning models. The system may determine, based on the knowledge graph and the query information, one or more correlation metrics indicating correlations between the machine-learning model and one or more features of the features in the knowledge graph. The system may determine one or more recommended features for the machine-learning model based on the one or more correlation metrics and the one or more features.