公开(公告)号：US20190335019A1

公开(公告)日：2019-10-31

申请号：US15962502

申请日：2018-04-25

Applicant: Google LLC

Inventor： Eugene Vladimir Davydov ,  Patrick Hummel ,  Jean-Francois Crespo ,  Shaohua Sun ,  Christopher Davis Monkman ,  Derek Leslie-Cook

IPC: H04L29/06

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining an adjusted requirement for transmission of a given digital component. In one aspect, a system includes a damping subsystem that obtains, from an evaluation subsystem, a standard requirement for transmission of the given digital component. The damping subsystem also obtains, from a prediction subsystem, a predicted requirement for transmission of the given digital component. The damping subsystem determines whether a damping condition is met. When the damping condition is met, the damping subsystem determines the adjusted requirement based on at least the predicted requirement. When the damping condition is not met, the damping subsystem determines the adjusted requirement based on at least the standard requirement.

公开(公告)号：US20210073638A1

公开(公告)日：2021-03-11

申请号：US17099387

申请日：2020-11-16

Applicant: Google LLC

Inventor： Benjamin Kenneth Coppin ,  Mustafa Suleyman ,  Thomas Chadwick Walters ,  Timothy Mann ,  Chia-Yueh Carlton Chu ,  Martin Szummer ,  Luis Carlos Cobo Rus ,  Jean-Francois Crespo

IPC: G06N3/08 ,  G06N20/00 ,  G06F16/9535 ,  G06F16/2457 ,  G06N3/04 ,  G06N7/00

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using a machine learning model that has been trained through reinforcement learning to select a content item. One of the methods includes receiving first data characterizing a first context in which a first content item may be presented to a first user in a presentation environment; and providing the first data as input to a long-term engagement machine learning model, the model having been trained through reinforcement learning to: receive a plurality of inputs, and process each of the plurality of inputs to generate a respective engagement score for each input that represents a predicted, time-adjusted total number of selections by the respective user of future content items presented to the respective user in the presentation environment if the respective content item is presented in the respective context.