Patents by Inventor CORNELIA CARAPCEA
CORNELIA CARAPCEA 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: 11829848Abstract: A method includes obtaining training data for a classifier, the training data comprises one or more target classes, obtaining candidate background classes, selecting negative classes from the candidate background classes, wherein the negative classes exclude candidate background classes that are close to the target classes, wherein the negative classes exclude candidate background classes that are very different from the target classes, and wherein the negative classes include candidate background classes that are similar to the target classes, and training the classifier on a combined set of the selected negative classes and target classes.Type: GrantFiled: June 8, 2017Date of Patent: November 28, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Yuxiao Hu, Lei Zhang, Christopher J Buehler, Anna Roth, Cornelia Carapcea
-
Patent number: 10691981Abstract: Methods, systems, and computer programs are presented for training a deep neural network (DNN). One method includes an operation for training a predecessor network defined for image recognition of items, where parameters of a predecessor classifier are initialized with random numbers sampled from a predetermined distribution, and the predecessor classifier utilizes an image-classification probability function without bias. The method further includes an operation for training a successor network defined for image recognition of items in a plurality of classes, where parameters of a successor classifier are initialized with parameters learned from the predecessor network, and the successor classifier utilizes the image-classification probability function without bias. Further, the method includes operations for receiving an image for recognition, and recognizing the image utilizing the successor classifier.Type: GrantFiled: March 11, 2019Date of Patent: June 23, 2020Assignee: Microsoft Technology Licensing, LLCInventors: Yandong Guo, Yuxiao Hu, Christopher J Buehler, Cornelia Carapcea, Lei Zhang
-
Publication number: 20190205705Abstract: Methods, systems, and computer programs are presented for training a deep neural network (DNN). One method includes an operation for training a predecessor network defined for image recognition of items, where parameters of a predecessor classifier are initialized with random numbers sampled from a predetermined distribution, and the predecessor classifier utilizes an image-classification probability function without bias. The method further includes an operation for training a successor network defined for image recognition of items in a plurality of classes, where parameters of a successor classifier are initialized with parameters learned from the predecessor network, and the successor classifier utilizes the image-classification probability function without bias. Further, the method includes operations for receiving an image for recognition, and recognizing the image utilizing the successor classifier.Type: ApplicationFiled: March 11, 2019Publication date: July 4, 2019Inventors: Yandong Guo, Yuxiao Hu, Christopher J. Buehler, Cornelia Carapcea, Lei Zhang
-
Patent number: 10262240Abstract: Methods, systems, and computer programs are presented for training a deep neural network (DNN). One method includes an operation for training a predecessor network defined for image recognition of items, where parameters of a predecessor classifier are initialized with random numbers sampled from a predetermined distribution, and the predecessor classifier utilizes an image-classification probability function without bias. The method further includes an operation for training a successor network defined for image recognition of items in a plurality of classes, where parameters of a successor classifier are initialized with parameters learned from the predecessor network, and the successor classifier utilizes the image-classification probability function without bias. Further, the method includes operations for receiving an image for recognition, and recognizing the image utilizing the successor classifier.Type: GrantFiled: August 14, 2017Date of Patent: April 16, 2019Assignee: Microsoft Technology Licensing, LLCInventors: Yandong Guo, Yuxiao Hu, Christopher J Buehler, Cornelia Carapcea, Lei Zhang
-
Publication number: 20190050689Abstract: Methods, systems, and computer programs are presented for training a deep neural network (DNN). One method includes an operation for training a predecessor network defined for image recognition of items, where parameters of a predecessor classifier are initialized with random numbers sampled from a predetermined distribution, and the predecessor classifier utilizes an image-classification probability function without bias. The method further includes an operation for training a successor network defined for image recognition of items in a plurality of classes, where parameters of a successor classifier are initialized with parameters learned from the predecessor network, and the successor classifier utilizes the image-classification probability function without bias. Further, the method includes operations for receiving an image for recognition, and recognizing the image utilizing the successor classifier.Type: ApplicationFiled: August 14, 2017Publication date: February 14, 2019Inventors: Yandong Guo, Yuxiao Hu, Christopher J Buehler, Cornelia Carapcea, Lei Zhang
-
Publication number: 20180330273Abstract: A method includes obtaining training data for a classifier, the training data comprises one or more target classes, obtaining candidate background classes, selecting negative classes from the candidate background classes, wherein the negative classes exclude candidate background classes that are close to the target classes, wherein the negative classes exclude candidate background classes that are very different from the target classes, and wherein the negative classes include candidate background classes that are similar to the target classes, and training the classifier on a combined set of the selected negative classes and target classes.Type: ApplicationFiled: June 8, 2017Publication date: November 15, 2018Inventors: YUXIAO HU, LEI ZHANG, Christopher J Buehler, ANNA ROTH, CORNELIA CARAPCEA
-
Publication number: 20180330272Abstract: A method includes obtaining a first classifier trained on a first dataset having a first dataset class, the first classifier having a plurality of first parameters, obtaining a second dataset having a second dataset class, loading the first parameters into a second classifier, merging a subset of the first dataset class and the second dataset class into a merged class, and training the second classifier using the merged class.Type: ApplicationFiled: June 7, 2017Publication date: November 15, 2018Inventors: Yuxiao Hu, Lei Zhang, Christopher Buehler, Cha Zhang, Anna Roth, Cornelia Carapcea
-
Patent number: 8983995Abstract: Systems, methods and computer-storage media are provided for identifying query formulation suggestions in response to receiving a search query. A portion of a search query is received. Query formulation suggestions are identified by semantically analyzing the search query. The query formulation suggestions are used to further formulate the received search query. The query formulation suggestions include semantic-pattern-based query suggestions that are derived from semantic query patterns, one or more entities, and information associated with these entities. The query formulation suggestions are transmitted for presentation.Type: GrantFiled: June 23, 2011Date of Patent: March 17, 2015Assignee: Microsoft CorporationInventors: Bo-June Hsu, Kuansan Wang, Yu-Ting Kuo, Chao-Chia Liu, Heung-Yeung Shum, Cornelia Carapcea, Yusuf Furkan Fidan, Lawrence William Colagiovanni, Arun Sacheti
-
Publication number: 20120265779Abstract: Systems, methods and computer-storage media are provided for identifying query formulation suggestions in response to receiving a search query. A portion of a search query is received. Query formulation suggestions are identified by semantically analyzing the search query. The query formulation suggestions are used to further formulate the received search query. The query formulation suggestions include semantic-pattern-based query suggestions that are derived from semantic query patterns, one or more entities, and information associated with these entities. The query formulation suggestions are transmitted for presentation.Type: ApplicationFiled: June 23, 2011Publication date: October 18, 2012Applicant: MICROSOFT CORPORATIONInventors: BO-JUNE HSU, KUANSAN WANG, YU-TING KUO, CHAO-CHIA LIU, HEUNG-YEUNG SHUM, CORNELIA CARAPCEA, YUSUF FURKAN FIDAN, LAWRENCE WILLIAM COLAGIOVANNI, ARUN SACHETI