Patents by Inventor Victor ZHONG
Victor ZHONG has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20240079784Abstract: An electronic device may be provided with an antenna having a resonating element and a light source module mounted to a flexible printed circuit and a metal cowling. The module may emit light through a rear housing wall. The printed circuit may be interposed between the metal cowling and a conductive support plate in the rear housing wall. The printed circuit may include a ground trace coupled to the resonating element. A dimpled pad may couple the ground trace to the support plate. Compressive foam may be used to exert a force against the flexible printed circuit that presses the dimpled pad against the conductive support plate. The ground trace and the dimpled pad may form a return path to ground for the resonating element. The dimpled pad may occupy less height within the device than other structures such as metal springs.Type: ApplicationFiled: August 30, 2023Publication date: March 7, 2024Inventors: Han Wang, Victor C. Lee, Jingni Zhong, Ming Chen, Bhaskara R. Rupakula, Yiren Wang, Yuan Tao, Christopher Q. Ma, Zhiheng Zhou, Sherry Cao, Kevin M. Froese, Hao Xu, Hongfei Hu, Mattia Pascolini
-
Patent number: 11836451Abstract: A method for maintaining a dialogue state associated with a dialogue between a user and a digital system includes receiving, by a dialogue state tracker associated with the digital system, a representation of a user communication, updating, by the dialogue state tracker, the dialogue state and providing a system response based on the updated dialogue state. The dialogue state is updated by evaluating, based on the representation of the user communication, a plurality of member scores corresponding to a plurality of ontology members of an ontology set, and selecting, based on the plurality of member scores, zero or more of the plurality of ontology members to add to or remove from the dialogue state.Type: GrantFiled: February 19, 2021Date of Patent: December 5, 2023Assignee: salesforce.com, inc.Inventors: Victor Zhong, Caiming Xiong
-
Patent number: 11526507Abstract: A computing system uses neural networks to translate natural language queries to database queries. The computing system uses a plurality of machine learning based models, each machine learning model for generating a portion of the database query. The machine learning models use an input representation generated based on terms of the input natural language query, a set of columns of the database schema, and the vocabulary of a database query language, for example, structured query language SQL. The plurality of machine learning based models may include an aggregation classifier model for determining an aggregation operator in the database query, a result column predictor model for determining the result columns of the database query, and a condition clause predictor model for determining the condition clause of the database query. The condition clause predictor is based on reinforcement learning.Type: GrantFiled: June 5, 2020Date of Patent: December 13, 2022Assignee: Salesforce, Inc.Inventors: Victor Zhong, Caiming Xiong, Richard Socher
-
Publication number: 20220044093Abstract: A computer-implemented method for dual sequence inference using a neural network model includes generating a codependent representation based on a first input representation of a first sequence and a second input representation of a second sequence using an encoder of the neural network model and generating an inference based on the codependent representation using a decoder of the neural network model. The neural network model includes a plurality of model parameters learned according to a machine learning process. The encoder includes a plurality of coattention layers arranged sequentially, each coattention layer being configured to receive a pair of layer input representations and generate one or more summary representations, and an output layer configured to receive the one or more summary representations from a last layer among the plurality of coattention layers and generate the codependent representation.Type: ApplicationFiled: October 20, 2021Publication date: February 10, 2022Inventors: Victor Zhong, Caiming Xiong, Richard Socher
-
Patent number: 11227218Abstract: A natural language processing system that includes a sentence selector and a question answering module. The sentence selector receives a question and sentences that are associated with a context. For a question and each sentence, the sentence selector determines a score. A score represents whether the question is answerable with the sentence. Sentence selector then generates a minimum set of sentences from the scores associated with the question and sentences. The question answering module generates an answer for the question from the minimum set of sentences.Type: GrantFiled: May 15, 2018Date of Patent: January 18, 2022Assignee: salesforce.com, inc.Inventors: Sewon Min, Victor Zhong, Caiming Xiong, Richard Socher
-
Patent number: 11170287Abstract: A computer-implemented method for dual sequence inference using a neural network model includes generating a codependent representation based on a first input representation of a first sequence and a second input representation of a second sequence using an encoder of the neural network model and generating an inference based on the codependent representation using a decoder of the neural network model. The neural network model includes a plurality of model parameters learned according to a machine learning process. The encoder includes a plurality of coattention layers arranged sequentially, each coattention layer being configured to receive a pair of layer input representations and generate one or more summary representations, and an output layer configured to receive the one or more summary representations from a last layer among the plurality of coattention layers and generate the codependent representation.Type: GrantFiled: January 26, 2018Date of Patent: November 9, 2021Assignee: salesforce.com, inc.Inventors: Victor Zhong, Caiming Xiong, Richard Socher
-
Publication number: 20210174028Abstract: A method for maintaining a dialogue state associated with a dialogue between a user and a digital system includes receiving, by a dialogue state tracker associated with the digital system, a representation of a user communication, updating, by the dialogue state tracker, the dialogue state and providing a system response based on the updated dialogue state. The dialogue state is updated by evaluating, based on the representation of the user communication, a plurality of member scores corresponding to a plurality of ontology members of an ontology set, and selecting, based on the plurality of member scores, zero or more of the plurality of ontology members to add to or remove from the dialogue state.Type: ApplicationFiled: February 19, 2021Publication date: June 10, 2021Inventors: Victor Zhong, Caiming Xiong
-
Patent number: 10963782Abstract: The technology disclosed relates to an end-to-end neural network for question answering, referred to herein as “dynamic coattention network (DCN)”. Roughly described, the DCN includes an encoder neural network and a coattentive encoder that capture the interactions between a question and a document in a so-called “coattention encoding”. The DCN also includes a decoder neural network and highway maxout networks that process the coattention encoding to estimate start and end positions of a phrase in the document that responds to the question.Type: GrantFiled: January 31, 2017Date of Patent: March 30, 2021Assignee: salesforce.com, inc.Inventors: Caiming Xiong, Victor Zhong, Richard Socher
-
Patent number: 10929607Abstract: A method for maintaining a dialogue state associated with a dialogue between a user and a digital system includes receiving, by a dialogue state tracker associated with the digital system, a representation of a user communication, updating, by the dialogue state tracker, the dialogue state and providing a system response based on the updated dialogue state. The dialogue state is updated by evaluating, based on the representation of the user communication, a plurality of member scores corresponding to a plurality of ontology members of an ontology set, and selecting, based on the plurality of member scores, zero or more of the plurality of ontology members to add to or remove from the dialogue state.Type: GrantFiled: May 14, 2018Date of Patent: February 23, 2021Assignee: salesforce.com, inc.Inventors: Victor Zhong, Caiming Xiong
-
Publication number: 20200301925Abstract: A computing system uses neural networks to translate natural language queries to database queries. The computing system uses a plurality of machine learning based models, each machine learning model for generating a portion of the database query. The machine learning models use an input representation generated based on terms of the input natural language query, a set of columns of the database schema, and the vocabulary of a database query language, for example, structured query language SQL. The plurality of machine learning based models may include an aggregation classifier model for determining an aggregation operator in the database query, a result column predictor model for determining the result columns of the database query, and a condition clause predictor model for determining the condition clause of the database query. The condition clause predictor is based on reinforcement learning.Type: ApplicationFiled: June 5, 2020Publication date: September 24, 2020Inventors: Victor Zhong, Caiming Xiong, Richard Socher
-
Patent number: 10747761Abstract: A computing system uses neural networks to translate natural language queries to database queries. The computing system uses a plurality of machine learning based models, each machine learning model for generating a portion of the database query. The machine learning models use an input representation generated based on terms of the input natural language query, a set of columns of the database schema, and the vocabulary of a database query language, for example, structured query language SQL. The plurality of machine learning based models may include an aggregation classifier model for determining an aggregation operator in the database query, a result column predictor model for determining the result columns of the database query, and a condition clause predictor model for determining the condition clause of the database query. The condition clause predictor is based on reinforcement learning.Type: GrantFiled: January 31, 2018Date of Patent: August 18, 2020Assignee: salesforce.com, inc.Inventors: Victor Zhong, Caiming Xiong, Richard Socher
-
Publication number: 20190258939Abstract: A natural language processing system that includes a sentence selector and a question answering module. The sentence selector receives a question and sentences that are associated with a context. For a question and each sentence, the sentence selector determines a score. A score represents whether the question is answerable with the sentence. Sentence selector then generates a minimum set of sentences from the scores associated with the question and sentences. The question answering module generates an answer for the question from the minimum set of sentences.Type: ApplicationFiled: May 15, 2018Publication date: August 22, 2019Inventors: Sewon Min, Victor Zhong, Caiming Xiong, Richard Socher
-
Publication number: 20190258714Abstract: A method for maintaining a dialogue state associated with a dialogue between a user and a digital system includes receiving, by a dialogue state tracker associated with the digital system, a representation of a user communication, updating, by the dialogue state tracker, the dialogue state and providing a system response based on the updated dialogue state. The dialogue state is updated by evaluating, based on the representation of the user communication, a plurality of member scores corresponding to a plurality of ontology members of an ontology set, and selecting, based on the plurality of member scores, zero or more of the plurality of ontology members to add to or remove from the dialogue state.Type: ApplicationFiled: May 14, 2018Publication date: August 22, 2019Inventors: Victor Zhong, Caiming Xiong
-
Publication number: 20190130248Abstract: A computer-implemented method for dual sequence inference using a neural network model includes generating a codependent representation based on a first input representation of a first sequence and a second input representation of a second sequence using an encoder of the neural network model and generating an inference based on the codependent representation using a decoder of the neural network model. The neural network model includes a plurality of model parameters learned according to a machine learning process. The encoder includes a plurality of coattention layers arranged sequentially, each coattention layer being configured to receive a pair of layer input representations and generate one or more summary representations, and an output layer configured to receive the one or more summary representations from a last layer among the plurality of coattention layers and generate the codependent representation.Type: ApplicationFiled: January 26, 2018Publication date: May 2, 2019Inventors: Victor Zhong, Caiming XIONG, Richard SOCHER
-
Publication number: 20180336198Abstract: A computing system uses neural networks to translate natural language queries to database queries. The computing system uses a plurality of machine learning based models, each machine learning model for generating a portion of the database query. The machine learning models use an input representation generated based on terms of the input natural language query, a set of columns of the database schema, and the vocabulary of a database query language, for example, structured query language SQL. The plurality of machine learning based models may include an aggregation classifier model for determining an aggregation operator in the database query, a result column predictor model for determining the result columns of the database query, and a condition clause predictor model for determining the condition clause of the database query. The condition clause predictor is based on reinforcement learning.Type: ApplicationFiled: January 31, 2018Publication date: November 22, 2018Inventors: Victor Zhong, Caiming Xiong, Richard Socher
-
Publication number: 20180129938Abstract: The technology disclosed relates to an end-to-end neural network for question answering, referred to herein as “dynamic coattention network (DCN)”. Roughly described, the DCN includes an encoder neural network and a coattentive encoder that capture the interactions between a question and a document in a so-called “coattention encoding”. The DCN also includes a decoder neural network and highway maxout networks that process the coattention encoding to estimate start and end positions of a phrase in the document that responds to the question.Type: ApplicationFiled: January 31, 2017Publication date: May 10, 2018Applicant: salesforce.com, inc.Inventors: Caiming XIONG, Victor ZHONG, Richard SOCHER