Patents by Inventor Richard Zemel
Richard Zemel 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).
-
Patent number: 11501192Abstract: Techniques for use in connection with performing optimization using an objective function that maps elements in a first domain to values in a range. The techniques include using at least one computer hardware processor to perform: identifying a first point at which to evaluate the objective function at least in part by using an acquisition utility function and a probabilistic model of the objective function, wherein the probabilistic model depends on a non-linear one-to-one mapping of elements in the first domain to elements in a second domain; evaluating the objective function at the identified first point to obtain a corresponding first value of the objective function; and updating the probabilistic model of the objective function using the first value to obtain an updated probabilistic model of the objective function.Type: GrantFiled: September 4, 2018Date of Patent: November 15, 2022Assignees: President and Fellows of Harvard College, The Governing Council of the University of TorontoInventors: Ryan P. Adams, Roland Jasper Snoek, Kevin Swersky, Richard Zemel
-
Publication number: 20200027012Abstract: Techniques for use in connection with performing optimization using an objective function that maps elements in a first domain to values in a range. The techniques include using at least one computer hardware processor to perform: identifying a first point at which to evaluate the objective function at least in part by using an acquisition utility function and a probabilistic model of the objective function, wherein the probabilistic model depends on a non-linear one-to-one mapping of elements in the first domain to elements in a second domain; evaluating the objective function at the identified first point to obtain a corresponding first value of the objective function; and updating the probabilistic model of the objective function using the first value to obtain an updated probabilistic model of the objective function.Type: ApplicationFiled: September 4, 2018Publication date: January 23, 2020Applicants: President and Fellows of Harvard College, Governing Council of the Univ. of Toronto, TheInventors: Ryan P. Adams, Roland Jasper Snoek, Kevin Swersky, Richard Zemel
-
Patent number: 10074054Abstract: Techniques for use in connection with performing optimization using an objective function that maps elements in a first domain to values in a range. The techniques include using at least one computer hardware processor to perform: identifying a first point at which to evaluate the objective function at least in part by using an acquisition utility function and a probabilistic model of the objective function, wherein the probabilistic model depends on a non-linear one-to-one mapping of elements in the first domain to elements in a second domain; evaluating the objective function at the identified first point to obtain a corresponding first value of the objective function; and updating the probabilistic model of the objective function using the first value to obtain an updated probabilistic model of the objective function.Type: GrantFiled: May 30, 2014Date of Patent: September 11, 2018Assignees: President and Fellows of Harvard College, Governing Council of the Univ. of Toronto, TheInventors: Ryan P. Adams, Roland Jasper Snoek, Kevin Swersky, Richard Zemel
-
Publication number: 20140358831Abstract: Techniques for use in connection with performing optimization using an objective function that maps elements in a first domain to values in a range. The techniques include using at least one computer hardware processor to perform: identifying a first point at which to evaluate the objective function at least in part by using an acquisition utility function and a probabilistic model of the objective function, wherein the probabilistic model depends on a non-linear one-to-one mapping of elements in the first domain to elements in a second domain; evaluating the objective function at the identified first point to obtain a corresponding first value of the objective function; and updating the probabilistic model of the objective function using the first value to obtain an updated probabilistic model of the objective function.Type: ApplicationFiled: May 30, 2014Publication date: December 4, 2014Applicants: President and Fellows of Harvard College, Governing Council of the Univ. of Toronto, The MaRS CentreInventors: Ryan P. Adams, Roland Jasper Snoek, Kevin Swersky, Richard Zemel
-
Patent number: 8027541Abstract: A system for organizing images includes an extraction component that extracts visual information (e.g., faces, scenes, etc.) from the images. The extracted visual information is provided to a comparison component which computes similarity confidence data between the extracted visual information. The similarity confidence data is an indication of the likelihood that items of extracted visual information are similar. The comparison component then generates a visual distribution of the extracted visual information based upon the similarity confidence data. The visual distribution can include groupings of the extracted visual information based on computed similarity confidence data. For example, the visual distribution can be a two-dimensional layout of faces organized based on the computed similarity confidence data—with faces in closer proximity faces computed to have a greater probability of representing the same person.Type: GrantFiled: March 15, 2007Date of Patent: September 27, 2011Assignee: Microsoft CorporationInventors: Gang Hua, Steven M. Drucker, Michael Revow, Paul A. Viola, Richard Zemel
-
Publication number: 20080226174Abstract: A system for organizing images includes an extraction component that extracts visual information (e.g., faces, scenes, etc.) from the images. The extracted visual information is provided to a comparison component which computes similarity confidence data between the extracted visual information. The similarity confidence data is an indication of the likelihood that items of extracted visual information are similar. The comparison component then generates a visual distribution of the extracted visual information based upon the similarity confidence data. The visual distribution can include groupings of the extracted visual information based on computed similarity confidence data. For example, the visual distribution can be a two-dimensional layout of faces organized based on the computed similarity confidence data—with faces in closer proximity faces computed to have a greater probability of representing the same person.Type: ApplicationFiled: March 15, 2007Publication date: September 18, 2008Applicant: Microsoft CorporationInventors: Gang Hua, Steven M. Drucker, Michael Revow, Paul A. Viola, Richard Zemel
-
Patent number: 5900096Abstract: A method of transferring metal leaf, such as gold leaf, to a substrate is disclosed. The method includes the steps of creating a transfer graphic design. After the transfer graphic design is created, a pressure sensitive adhesive design in the shape of the transfer graphic design is formed on a transfer sheet having paper backing and a release layer. The transfer sheet may be a dry transfer sheet or a water release decal type transfer sheet. The transfer sheet containing the pressure sensitive adhesive design is then placed on a substrate to which the pressure sensitive adhesive design is transferred by removing the transfer sheet so that the pressure sensitive adhesive design adheres to the substrate. Metal leaf is then applied to the pressure sensitive adhesive design.Type: GrantFiled: September 3, 1996Date of Patent: May 4, 1999Inventor: Richard Zemel