公开(公告)号：US20250111272A1

公开(公告)日：2025-04-03

申请号：US18574668

申请日：2023-04-25

Applicant: Google LLC

Inventor： Wei Huang ,  Zhenyu Liu ,  Liang Wang ,  Kumar Rishabh

IPC: G06N20/00

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for distributing digital contents to client devices are described. The system obtains, for each user in a set of users, user attribute data and, for a subset of the users, consent data for controlling usage of the user attribute data. The system partitions, based at least on the consent data for the subset of users, the set of users into a first group of users and a second group of users. The system generates a respective training dataset based on the data for each group of user, and uses the datasets to train a machine learning model configured to predict information about one or more users. In particular, the system applies differential privacy to the second training dataset without applying differential privacy to the first training dataset during training.

公开(公告)号：US20250094508A1

公开(公告)日：2025-03-20

申请号：US18577448

申请日：2023-01-18

Applicant: Google LLC

Inventor： Wei Huang ,  Zhenyu Liu ,  Liang Wang ,  Kumar Rishabh

IPC: G06F16/9535 ,  G06F16/9537

Abstract: Methods, systems, and media comprising; obtaining, from a client device and during a browsing session conducted by a user, contextual features relating to context within the browsing session, wherein the contextual features do not include any personally-identifiable data; generating, using a trained contextual model and based on the contextual features, an audience interest profile, wherein the audience interest profile represents a prediction of affinity to one or more content categories, wherein the trained contextual model is trained using a set of historical contextual data aggregated from a plurality of prior browsing sessions and audience interest profiles that each represent an affinity to one or more content categories, and wherein the set of historical contextual data does not include any personally-identifiable data; identifying, based on the generated audience interest profile, a digital component for provision; and providing, for display on the client device and during the browsing session, the digital component.