公开(公告)号：US11800167B2

公开(公告)日：2023-10-24

申请号：US17515225

申请日：2021-10-29

Applicant: Roku, Inc.

Inventor： Amit Paliwal ,  Andrey Marsavin ,  Govind Vaidya ,  Wim Michiels ,  Beth Teresa Logan ,  Zheng Han ,  Tapan Oza ,  Vijay Anand Raghavan

IPC: H04N21/2662 ,  H04N21/24 ,  G06N20/00 ,  H04N21/25

CPC classification number: H04N21/2662 ,  G06N20/00 ,  H04N21/2401 ,  H04N21/251

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for modifying one or more parameters of a data streaming bitrate selection algorithm based on machine learning. An example embodiment operates by training and operating a first machine learning model to predict a sustainable network bandwidth. A second machine learning model is trained to receive the sustainable network bandwidth and predict a likelihood that this network bandwidth will not empty a data buffer of streaming data. A bitrate is selected based on the likelihood being below a threshold percentage, such as 50%.

公开(公告)号：US20230421831A1

公开(公告)日：2023-12-28

申请号：US18462635

申请日：2023-09-07

Applicant: ROKU, INC.

Inventor： Amit Paliwal ,  Andrey Marsavin ,  Govind Vaidya ,  Wim Michiels ,  Beth Teresa Logan ,  Zheng Han ,  Tapan Oza ,  Vijay Anand Raghavan

IPC: H04N21/2662 ,  H04N21/24 ,  G06N20/00 ,  H04N21/25

CPC classification number: H04N21/2662 ,  H04N21/2401 ,  G06N20/00 ,  H04N21/251

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for modifying one or more parameters of a data streaming bitrate selection algorithm based on machine learning. An example embodiment operates by training and operating a first machine learning model to predict a sustainable network bandwidth. A second machine learning model is trained to receive the sustainable network bandwidth and predict a likelihood that this network bandwidth will not empty a data buffer of streaming data. A bitrate is selected based on the likelihood being below a threshold percentage, such as 50%.

公开(公告)号：US12137265B2

公开(公告)日：2024-11-05

申请号：US18462635

申请日：2023-09-07

Applicant: ROKU, INC.

Inventor： Amit Paliwal ,  Andrey Marsavin ,  Govind Vaidya ,  Wim Michiels ,  Beth Teresa Logan ,  Zheng Han ,  Tapan Oza ,  Vijay Anand Raghavan

IPC: H04N21/2662 ,  G06N5/01 ,  G06N20/00 ,  G06N20/20 ,  H04N21/24 ,  H04N21/25 ,  H04N21/44 ,  H04N21/442 ,  H04N21/845

Abstract: Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for modifying one or more parameters of a data streaming bitrate selection algorithm based on machine learning. An example embodiment operates by training and operating a first machine learning model to predict a sustainable network bandwidth. A second machine learning model is trained to receive the sustainable network bandwidth and predict a likelihood that this network bandwidth will not empty a data buffer of streaming data. A bitrate is selected based on the likelihood being below a threshold percentage, such as 50%.