公开(公告)号：US20250028774A1

公开(公告)日：2025-01-23

申请号：US18778014

申请日：2024-07-19

Applicant: GOOGLE LLC

Inventor： Nasim Sedaghat ,  Esteban García Sánchez ,  Jorge Zuniga ,  Katharine Giari ,  Nicolas MacBeth ,  Sébastien Séguin-Gagnon ,  Samuel Birch ,  Ayman Almadhoun ,  Armina Foroughi-Shafiei ,  Rui Feng ,  Sara Hee Shin Park ,  Yue Zhang ,  Matthew Jones ,  Yu Liang Fang

IPC: G06F16/957 ,  G06F3/0483 ,  G06F16/955

Abstract: A method may determine that first content in a first tab of a browser is associated with a first entity of an entity type and determine that second content in a second tab of the browser is associated with a second entity of the entity type. Responsive to determining that the first tab is associated with the first entity of the entity type and that the second tab is associated with the second entity of the entity type, a method may generate information used to provide a user interface for displaying first information for the first entity and second information for the second entity.

公开(公告)号：US20240119366A1

公开(公告)日：2024-04-11

申请号：US18477525

申请日：2023-09-28

Applicant: Google LLC

Inventor： Matthew Jones ,  Michael Curtis Mozer

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for online training of machine learning models predicting time-series data. In one aspect, a method comprises training a machine learning model having a plurality of weights by maintaining weight data, specifying a plurality of sub-weights for each of the plurality of weights and covariance data that estimates the joint uncertainty between the sub-weights, and, at each of a plurality of time steps, receiving model inputs, processing the model inputs using the weight data to generate corresponding model outputs, receiving corresponding ground truth outputs, and updating the weight data using the corresponding ground truth outputs.