公开(公告)号：US20230377572A1

公开(公告)日：2023-11-23

申请号：US18231112

申请日：2023-08-07

Applicant: GOOGLE LLC

Inventor： Mugurel-Ionut Andreica ,  Vladimir Vuskovic ,  Joseph Lange ,  Sharon Stovezky ,  Marcin Nowak-Przygodzki

IPC: G10L15/22 ,  G06N3/08 ,  G10L15/02

CPC classification number: G10L15/22 ,  G06N3/08 ,  G10L15/02 ,  G10L2015/223

Abstract: Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

公开(公告)号：US11769502B2

公开(公告)日：2023-09-26

申请号：US17339114

申请日：2021-06-04

Applicant: Google LLC

Inventor： Mugurel Ionut Andreica ,  Vladimir Vuskovic ,  Joseph Lange ,  Sharon Stovezky ,  Marcin Nowak-Przygodzki

IPC: G10L15/22 ,  G06N3/08 ,  G10L15/02

CPC classification number: G10L15/22 ,  G06N3/08 ,  G10L15/02 ,  G10L2015/223

Abstract: Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

公开(公告)号：US20230177272A1

公开(公告)日：2023-06-08

申请号：US18103255

申请日：2023-01-30

Applicant: Google LLC

Inventor： Sharon Stovezky ,  Yariv Adan ,  Radu Voroneanu ,  Behshad Behzadi ,  Ragnar Groot Koerkamp ,  Marcin Nowak-Przygodzki

IPC: G06F40/295 ,  G06F16/9537

CPC classification number: G06F40/295 ,  G06F16/9537

Abstract: Implementations set forth herein relate to an automated assistant that operates according to a variety of different location-based biasing modes for rendering responsive content for a user and/or proactively suggesting content for the user. The user can provide condensed spoken utterances to the automated assistant, when the automated assistant is operating according to one or more location-based biasing modes, but nonetheless receive accurate responsive outputs from the automated assistant. A responsive output generated by biasing toward a subset of location characteristic data that has been prioritized over other subsets of location characteristic data. The biasing allows the automated assistant to compensate for any details that may be missing from a spoken utterance, but allows the user to provide shorter spoken utterances, thereby reducing an amount of language processing when processing inputs from the user.

公开(公告)号：US12118998B2

公开(公告)日：2024-10-15

申请号：US18231112

申请日：2023-08-07

Applicant: GOOGLE LLC

Inventor： Mugurel Ionut Andreica ,  Vladimir Vuskovic ,  Joseph Lange ,  Sharon Stovezky ,  Marcin Nowak-Przygodzki

IPC: G10L15/22 ,  G06N3/08 ,  G10L15/02

CPC classification number: G10L15/22 ,  G06N3/08 ,  G10L15/02 ,  G10L2015/223

Abstract: Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

公开(公告)号：US11568146B2

公开(公告)日：2023-01-31

申请号：US16605838

申请日：2019-09-10

Applicant: Google LLC

Inventor： Sharon Stovezky ,  Yariv Adan ,  Radu Voroneanu ,  Behshad Behzadi ,  Ragnar Groot Koerkamp ,  Marcin Nowak-Przygodzki

IPC: G06F40/295 ,  G06F16/9537

Abstract: Implementations set forth herein relate to an automated assistant that operates according to a variety of different location-based biasing modes for rendering responsive content for a user and/or proactively suggesting content for the user. The user can provide condensed spoken utterances to the automated assistant, when the automated assistant is operating according to one or more location-based biasing modes, but nonetheless receive accurate responsive outputs from the automated assistant. A responsive output generated by biasing toward a subset of location characteristic data that has been prioritized over other subsets of location characteristic data. The biasing allows the automated assistant to compensate for any details that may be missing from a spoken utterance, but allows the user to provide shorter spoken utterances, thereby reducing an amount of language processing when processing inputs from the user.

公开(公告)号：US20210295841A1

公开(公告)日：2021-09-23

申请号：US17339114

申请日：2021-06-04

Applicant: Google LLC

Inventor： Mugurel Ionut Andreica ,  Vladimir Vuskovic ,  Joseph Lange ,  Sharon Stovezky ,  Marcin Nowak-Przygodzki

IPC: G10L15/22 ,  G06N3/08 ,  G10L15/02

Abstract: Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

公开(公告)号：US11031007B2

公开(公告)日：2021-06-08

申请号：US16343285

申请日：2019-02-07

Applicant: Google LLC

Inventor： Mugurel Ionut Andreica ,  Vladimir Vuskovic ,  Joseph Lange ,  Sharon Stovezky ,  Marcin Nowak-Przygodzki

IPC: G10L15/22 ,  G06N3/08 ,  G10L15/02

Abstract: Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

公开(公告)号：US20210069986A1

公开(公告)日：2021-03-11

申请号：US16605838

申请日：2019-09-10

Applicant: Google LLC

Inventor： Sharon Stovezky ,  Yariv Adan ,  Radu Voroneanu ,  Behshad Behzadi ,  Ragnar Groot Koerkamp ,  Marcin Nowak-Przygodzki

IPC: B29C64/393 ,  B29C64/295 ,  B33Y70/00 ,  B33Y10/00 ,  B33Y30/00

Abstract: In various embodiments, the invention relates to poly(propylene fumarate) (PPF)-based star-shaped copolymers synthesized using a core-first approach that uses a multi-functional alcohols as an initiator, and Mg(BHT)2(THF)2 as catalyst for controlled ring opening copolymerization (ROCOP) of maleic anhydride (MAn) with propylene oxide (PO). In some embodiments, these star-PPF copolymers have lower viscosities than their linear analogs, allowing a decrease in DEF fraction in resin formulation, as well as the use of higher molecular weights. These star-shape PPF can be used to prepare PPF:DEF resins containing as much as 70% by weight of the multi-arm PPF star copolymers, and have a low complex viscosity of high Mn star PPF resin that affords rapid printing with a Mn nearly eight times larger than the largest linear PPF oligomer printed previously.

公开(公告)号：US20200302924A1

公开(公告)日：2020-09-24

申请号：US16343285

申请日：2019-02-07

Applicant: Google LLC

Inventor： Mugurel Ionut Andreica ,  Vladimir Vuskovic ,  Joseph Lange ,  Sharon Stovezky ,  Marcin Nowak-Przygodzki

IPC: G10L15/22 ,  G10L15/02 ,  G06N3/08

Abstract: Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.