Patents by Inventor Patrick F. Riley
Patrick F. Riley 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: 20240006027Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for obtaining data defining a sequence for an aptamer, the aptamer comprising a string of nucleobases; encoding the data defining the sequence for the aptamer as a neural network input; and processing the neural network input using a neural network to generate an output that characterizes how strongly the aptamer binds to a particular target molecule, wherein the neural network has been configured through training to receive the data defining the sequence and to process the data to generate predicted outputs that characterize how strongly the aptamer binds to the particular target molecule.Type: ApplicationFiled: April 26, 2023Publication date: January 4, 2024Inventors: Michelle Therese Hoerner Dimon, Marc Berndl, Marc Adlai Coram, Brian Trippe, Patrick F. Riley, Philip Charles Nelson
-
Publication number: 20230334293Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: ApplicationFiled: April 19, 2023Publication date: October 19, 2023Inventors: Patrick F. Riley, Marc Berndl
-
Patent number: 11670400Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for obtaining data defining a sequence for an aptamer, the aptamer comprising a string of nucleobases; encoding the data defining the sequence for the aptamer as a neural network input; and processing the neural network input using a neural network to generate an output that characterizes how strongly the aptamer binds to a particular target molecule, wherein the neural network has been configured through training to receive the data defining the sequence and to process the data to generate predicted outputs that characterize how strongly the aptamer binds to the particular target molecule.Type: GrantFiled: January 24, 2020Date of Patent: June 6, 2023Assignee: Google LLCInventors: Michelle Therese Hoerner Dimon, Marc Berndl, Marc Adlai Coram, Brian Trippe, Patrick F. Riley, Philip Charles Nelson
-
Patent number: 11663447Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: GrantFiled: August 24, 2021Date of Patent: May 30, 2023Assignee: Google LLCInventors: Patrick F. Riley, Marc Berndl
-
Patent number: 11205113Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: GrantFiled: July 29, 2019Date of Patent: December 21, 2021Assignee: Google LLCInventors: Patrick F. Riley, Marc Berndl
-
Publication number: 20210383196Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: ApplicationFiled: August 24, 2021Publication date: December 9, 2021Inventors: Patrick F. Riley, Marc Berndl
-
Publication number: 20200160937Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for obtaining data defining a sequence for an aptamer, the aptamer comprising a string of nucleobases; encoding the data defining the sequence for the aptamer as a neural network input; and processing the neural network input using a neural network to generate an output that characterizes how strongly the aptamer binds to a particular target molecule, wherein the neural network has been configured through training to receive the data defining the sequence and to process the data to generate predicted outputs that characterize how strongly the aptamer binds to the particular target molecule.Type: ApplicationFiled: January 24, 2020Publication date: May 21, 2020Inventors: Michelle Therese Hoerner Dimon, Marc Berndl, Marc Adlai Coram, Brian Trippe, Patrick F. Riley, Philip Charles Nelson
-
Patent number: 10546650Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for obtaining data defining a sequence for an aptamer, the aptamer comprising a string of nucleobases; encoding the data defining the sequence for the aptamer as a neural network input; and processing the neural network input using a neural network to generate an output that characterizes how strongly the aptamer binds to a particular target molecule, wherein the neural network has been configured through training to receive the data defining the sequence and to process the data to generate predicted outputs that characterize how strongly the aptamer binds to the particular target molecule.Type: GrantFiled: October 23, 2015Date of Patent: January 28, 2020Assignee: Google LLCInventors: Michelle Therese Hoerner Dimon, Marc Berndl, Marc Adlai Coram, Brian Trippe, Patrick F. Riley, Philip Charles Nelson
-
Publication number: 20190354840Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: ApplicationFiled: July 29, 2019Publication date: November 21, 2019Inventors: Patrick F. Riley, Marc Berndl
-
Patent number: 10366324Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: GrantFiled: September 1, 2015Date of Patent: July 30, 2019Assignee: Google LLCInventors: Patrick F. Riley, Marc Berndl
-
Publication number: 20170116371Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for obtaining data defining a sequence for an aptamer, the aptamer comprising a string of nucleobases; encoding the data defining the sequence for the aptamer as a neural network input; and processing the neural network input using a neural network to generate an output that characterizes how strongly the aptamer binds to a particular target molecule, wherein the neural network has been configured through training to receive the data defining the sequence and to process the data to generate predicted outputs that characterize how strongly the aptamer binds to the particular target molecule.Type: ApplicationFiled: October 23, 2015Publication date: April 27, 2017Applicant: GOOGLE INC.Inventors: Michelle Therese Hoerner Dimon, Marc Berndl, Marc Adlai Coram, Brian Trippe, Patrick F. Riley, Philip Charles Nelson
-
Publication number: 20170061276Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving graph data representing an input graph comprising a plurality of vertices connected by edges; generating, from the graph data, vertex input data representing characteristics of each vertex in the input graph and pair input data representing characteristics of pairs of vertices in the input graph; and generating order-invariant features of the input graph using a neural network, wherein the neural network comprises: a first subnetwork configured to generate a first alternative representation of the vertex input data and a first alternative representation of the pair input data from the vertex input data and the pair input data; and a combining layer configured to receive an input alternative representation and to process the input alternative representation to generate the order-invariant features.Type: ApplicationFiled: September 1, 2015Publication date: March 2, 2017Inventors: Patrick F. Riley, Marc Berndl
-
Patent number: 9323853Abstract: The present invention provides systems and methods for generating alerts based on results received from one or more custom search engines. In an embodiment, the present invention generates alerts based on custom search engines by (1) receiving one or more alert specifications including a custom search engine identifier; (2) receiving at least one current result from the identified custom search engines; (3) determining one or more itemized alert elements based on the at least one current result; and (4) transmitting a signal representing the one or more itemized alert elements to a user via a communication network.Type: GrantFiled: April 22, 2014Date of Patent: April 26, 2016Assignee: Google Inc.Inventors: Patrick F. Riley, Ramanathan V. Guha
-
Publication number: 20150331906Abstract: A system embodiment receives a query from the user. The query is processed by a search infrastructure which may return a set of results. Results may then be estimated for quality by a quality estimator. The quality estimator may determine an initial quality estimation of the search result(s). A query re-issuer generates re-issued query information based on the initial quality estimation of the search result(s) for the initial query. A query modifier then receives the re-issued query information and generates a modified re-issued query in accordance with the query re-issue information. A method embodiment modifies and re-issues a query to the search infrastructure. The method includes determining an initial quality estimation of the search result, generating re-issued query information based on at least an initial quality estimation of the search result for the initial query, receiving the re-issued query information and generating a modified re-issued query.Type: ApplicationFiled: July 30, 2015Publication date: November 19, 2015Inventors: Patrick F. Riley, Paul Komarek, Vrishali Wagle
-
Patent number: 9116993Abstract: A system embodiment receives a query from the user. The query is processed by a search infrastructure which may return a set of results. Results may then be estimated for quality by a quality estimator. The quality estimator may determine an initial quality estimation of the search result(s). A query re-issuer generates re-issued query information based on the initial quality estimation of the search result(s) for the initial query. A query modifier then receives the re-issued query information and generates a modified re-issued query in accordance with the query re-issue information. A method embodiment modifies and re-issues a query to the search infrastructure. The method includes determining an initial quality estimation of the search result, generating re-issued query information based on at least an initial quality estimation of the search result for the initial query, receiving the re-issued query information and generating a modified re-issued query.Type: GrantFiled: December 27, 2013Date of Patent: August 25, 2015Assignee: Google Inc.Inventors: Patrick F. Riley, Paul Komarek, Vrishali Wagle
-
Patent number: 9043322Abstract: This invention relates to determining page elements to display in response to a search. A method embodiment of this invention determines a page element based on a search result. The method includes: (1) determining a set of result classifications based on the search result, wherein each result classification includes a result category and a result score; and (2) determining the page element based on the set of result classifications. In this way, a classification is determined based on a search result and page elements are generated based on the classification. By using the search result, as opposed to just the query, page elements are generated that corresponds to a predominant interpretation of the user's query within the search results. As result, the page elements may, in most cases, accurately reflect the user's intent.Type: GrantFiled: November 25, 2013Date of Patent: May 26, 2015Assignee: Google Inc.Inventors: Tania Bedrax-Weiss, Ramanathan V. Guha, Patrick F. Riley, Corin Anderson
-
Patent number: 8909619Abstract: The subject matter of this specification can be embodied in, among other things, a method that includes obtaining, in response to a determination that one or more tools are relevant to a search query or search results, code for presenting user interface controls with the search results, wherein selection of a particular presented user interface control results in performing an operation corresponding to a particular tool for manipulating the search results, and wherein the code is configured to present the user interface controls for the determined tools more prominently than user interface controls for one or more others of the tools that were not determined to be relevant to the search query or the search results.Type: GrantFiled: February 3, 2011Date of Patent: December 9, 2014Assignee: Google Inc.Inventors: Patrick F. Riley, Yujing Zeng, Dale K. Neal
-
Patent number: 8892552Abstract: A user can issue a query on a search host site. The search host can construct a corresponding reference to custom search engine (CSE) with a URL pointing to a site with information for a CSE specification. The search host sends the reference to a CSE provider and the CSE provider dynamically generates the CSE specification based on the reference. The CSE provider then conducts a search and processes search results accordingly. The processed search results are sent to the search host who outputs the search results for the user. The information for the CSE specification can be stored anywhere accessible in a computer network and can be stored in various formats. Any number of CSE specifications can be automatically generated on demand corresponding to different queries or content requests on the search host site.Type: GrantFiled: March 11, 2008Date of Patent: November 18, 2014Assignee: Google Inc.Inventors: Othar Hansson, Matthew D. Wytock, Patrick F. Riley, Ramanathan V. Guha
-
Patent number: 8725716Abstract: The present invention provides systems and methods for generating alerts based on results received from one or more custom search engines. In an embodiment, the present invention generates alerts based on custom search engines by (1) receiving one or more alert specifications including a custom search engine identifier; (2) receiving at least one current result from the identified custom search engines; (3) determining one or more itemized alert elements based on the at least one current result; and (4) transmitting a signal representing the one or more itemized alert elements to a user via a communication network.Type: GrantFiled: May 30, 2008Date of Patent: May 13, 2014Assignee: Google Inc.Inventors: Patrick F. Riley, Ramanathan V. Guha
-
Publication number: 20140114945Abstract: A system embodiment receives a query from the user. The query is processed by a search infrastructure which may return a set of results. Results may then be estimated for quality by a quality estimator. The quality estimator may determine an initial quality estimation of the search result(s). A query re-issuer generates re-issued query information based on the initial quality estimation of the search result(s) for the initial query. A query modifier then receives the re-issued query information and generates a modified re-issued query in accordance with the query re-issue information. A method embodiment modifies and re-issues a query to the search infrastructure. The method includes determining an initial quality estimation of the search result, generating re-issued query information based on at least an initial quality estimation of the search result for the initial query, receiving the re-issued query information and generating a modified re-issued query.Type: ApplicationFiled: December 27, 2013Publication date: April 24, 2014Applicant: Google Inc.Inventors: Patrick F. Riley, Paul Komarek, Vrishali Wagle