Patents by Inventor Shengjun Pan
Shengjun Pan 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: 20230289662Abstract: Shaded bid values may be determined and/or submitted to one or more auction modules for participation in auctions. Auction information including at least one of impression indications associated with the auctions, sets of features associated with the auctions, the shaded bid values associated with the auctions, etc. may be stored in a database. A machine learning model may be trained using the auction information to generate a first machine learning model with feature parameters associated with features. A bid request, indicative of a second set of features, may be received. The first machine learning model may be used to determine win probabilities and/or expected bid surpluses associated with multiple shaded bid values based upon one or more feature parameters, of the feature parameters, associated with the second set of features. A shaded bid value for submission may be determined based upon the win probabilities and/or the expected bid surpluses.Type: ApplicationFiled: May 21, 2023Publication date: September 14, 2023Inventors: Shengjun Pan, Tian Zhou, Brendan Kitts, Hao He, Bharatbhushan Shetty, Djordje Gligorijevic, Junwei Pan, Tingyu Mao, San Guitekin, Balaji Srinivasa Rao Paladugu, Jianlong Zhang, Sneha Thomas, Aaron Flores
-
Publication number: 20230281512Abstract: One or more computing devices, systems, and/or methods are provided. Shaded bid values may be determined and/or submitted to one or more auction modules for participation in auctions. Auction information including at least one of minimum bid values to win associated with the auctions, sets of features associated with the auctions, the shaded bid values associated with the auctions, unshaded bid values associated with the auctions, etc. may be stored in a database. A machine learning model may be trained using a loss function and/or the auction information to generate a first machine learning model with feature parameters associated with features. A bid request, indicative of a second set of features, may be received. The first machine learning model may be used to determine a shaded bid value for submission based upon one or more first feature parameters, of the feature parameters, associated with the second set of features.Type: ApplicationFiled: May 15, 2023Publication date: September 7, 2023Inventors: Tian Zhou, Djoefje Gligorijevic, Bharatbhushan Shetty, Junwei Pan, Brendan kitts, Shengjun Pan, Balaji Srinivasa Rao Paladugu, Sneha Thomas, Aaron Flores
-
Patent number: 11657326Abstract: Shaded bid values may be determined and/or submitted to one or more auction modules for participation in auctions. Auction information including at least one of impression indications associated with the auctions, sets of features associated with the auctions, the shaded bid values associated with the auctions, etc. may be stored in a database. A machine learning model may be trained using the auction information to generate a first machine learning model with feature parameters associated with features. A bid request, indicative of a second set of features, may be received. The first machine learning model may be used to determine win probabilities and/or expected bid surpluses associated with multiple shaded bid values based upon one or more feature parameters, of the feature parameters, associated with the second set of features. A shaded bid value for submission may be determined based upon the win probabilities and/or the expected bid surpluses.Type: GrantFiled: August 17, 2020Date of Patent: May 23, 2023Assignee: YAHOO AD TECH LLCInventors: Shengjun Pan, Tian Zhou, Brendan Kitts, Hao He, Bharatbhushan Shetty, Djordje Gligorijevic, Junwei Pan, Tingyu Mao, San Gultekin, Balaji Srinivasa Rao Paladugu, Jianlong Zhang, Sneha Thomas, Aaron Flores
-
Patent number: 11651284Abstract: One or more computing devices, systems, and/or methods are provided. Shaded bid values may be determined and/or submitted to one or more auction modules for participation in auctions. Auction information including at least one of minimum bid values to win associated with the auctions, sets of features associated with the auctions, the shaded bid values associated with the auctions, unshaded bid values associated with the auctions, etc. may be stored in a database. A machine learning model may be trained using a loss function and/or the auction information to generate a first machine learning model with feature parameters associated with features. A bid request, indicative of a second set of features, may be received. The first machine learning model may be used to determine a shaded bid value for submission based upon one or more first feature parameters, of the feature parameters, associated with the second set of features.Type: GrantFiled: August 17, 2020Date of Patent: May 16, 2023Assignee: YAHOO AD TECH LLCInventors: Tian Zhou, Djordje Gligorijevic, Bharatbhushan Shetty, Junwei Pan, Brendan Kitts, Shengjun Pan, Balaji Srinivasa Rao Paladugu, Sneha Thomas, Aaron Flores
-
Publication number: 20220383168Abstract: The present teaching generally relates to removing perturbations from predictive scoring. In one embodiment, data representing a plurality of events detected by a content provider may be received, the data indicating a time that a corresponding event occurred and whether the corresponding event was fraudulent. First category data may be generated by grouping each event into one of a number of categories, each category being associated with a range of times. A first measure of risk for each category may be determined, where the first measure of risk indicates a likelihood that a future event occurring at a future time is fraudulent. Second category data may be generated by processing the first category data and a second measure of risk for each category may be determined. Measure data representing the second measure of risk for each category and the range of times associated with that category may be stored.Type: ApplicationFiled: August 5, 2022Publication date: December 1, 2022Inventors: Liang Wang, Angus Xianen Qiu, Shengjun Pan
-
Patent number: 11410062Abstract: The present teaching generally relates to removing perturbations from predictive scoring. In one embodiment, data representing a plurality of events detected by a content provider may be received, the data indicating a time that a corresponding event occurred and whether the corresponding event was fraudulent. First category data may be generated by grouping each event into one of a number of categories, each category being associated with a range of times. A first measure of risk for each category may be determined, where the first measure of risk indicates a likelihood that a future event occurring at a future time is fraudulent. Second category data may be generated by processing the first category data and a second measure of risk for each category may be determined. Measure data representing the second measure of risk for each category and the range of times associated with that category may be stored.Type: GrantFiled: December 19, 2017Date of Patent: August 9, 2022Assignee: YAHOO AD TECH LLCInventors: Liang Wang, Angus Xianen Qiu, Shengjun Pan
-
Publication number: 20220051130Abstract: One or more computing devices, systems, and/or methods are provided. Shaded bid values may be determined and/or submitted to one or more auction modules for participation in auctions. Auction information including at least one of minimum bid values to win associated with the auctions, sets of features associated with the auctions, the shaded bid values associated with the auctions, unshaded bid values associated with the auctions, etc. may be stored in a database. A machine learning model may be trained using a loss function and/or the auction information to generate a first machine learning model with feature parameters associated with features. A bid request, indicative of a second set of features, may be received. The first machine learning model may be used to determine a shaded bid value for submission based upon one or more first feature parameters, of the feature parameters, associated with the second set of features.Type: ApplicationFiled: August 17, 2020Publication date: February 17, 2022Inventors: Tian Zhou, Djordje Gligorijevic, Bharatbhushan Shetty, Junwei Pan, Brendan Kitts, Shengjun Pan, Balaji Srinivasa Rao Paladugu, Sneha Thomas, Aaron Flores
-
Publication number: 20220051131Abstract: Shaded bid values may be determined and/or submitted to one or more auction modules for participation in auctions. Auction information including at least one of impression indications associated with the auctions, sets of features associated with the auctions, the shaded bid values associated with the auctions, etc. may be stored in a database. A machine learning model may be trained using the auction information to generate a first machine learning model with feature parameters associated with features. A bid request, indicative of a second set of features, may be received. The first machine learning model may be used to determine win probabilities and/or expected bid surpluses associated with multiple shaded bid values based upon one or more feature parameters, of the feature parameters, associated with the second set of features. A shaded bid value for submission may be determined based upon the win probabilities and/or the expected bid surpluses.Type: ApplicationFiled: August 17, 2020Publication date: February 17, 2022Inventors: Shengjun Pan, Tian Zhou, Brendan Kitts, Hao He, Bharatbhushan Shetty, Djordje Gligorijevic, Junwei Pan, Tingyu Mao, San Gultekin, Balaji Srinivasa Rao Paladugu, Jianlong Zhang, Sneha Thomas, Aaron Flores
-
Publication number: 20210201501Abstract: A motion-based object detection method includes the steps of extracting, by processing acquired first and second images, one or more regions of interest (ROIs); transforming the one or more ROIs into grayscale; and acquiring, by processing the grayscale ROIs with a deep neural network (DNN) model to classify the objects contained in the one or more ROIs, a classification result of whether the objects contained in the one or more ROIs belong to a given categories. The DNN model comprises N (N is a positive integer and ranged from 4-12) depthwise separable convolution layers. each depthwise separable convolution layer comprises a depthwise convolution layer for applying a single filter to each input channel and a pointwise layer for creating a linear combination of the outputs of the depthwise convolution layer to obtain feature maps of the grayscale ROIs.Type: ApplicationFiled: June 29, 2018Publication date: July 1, 2021Inventors: Po YUAN, Shengjun PAN, Junneng ZHAO, Daniel MARINIUC
-
Patent number: 10657783Abstract: A video surveillance method includes the steps of detecting, by a motion detector, an object motion, in the field of view of a surveillance device when the surveillance device is in a standby mode; generating, responsive to a positive detection to switch the surveillance device from the standby mode to an operation mode, one or more images of the moving object; determining, by processing the one or more images with a deep neural network (DNN) model of an object detector, whether the objects contained in the one or more images belong to a given categories, wherein the DNN model comprises N (N is a positive integer and ranged from 4-12) depthwise separable convolution layers; and video recording, responsive to a positive determination, the moving object in the field of view of the surveillance device.Type: GrantFiled: January 2, 2019Date of Patent: May 19, 2020Assignee: Hangzhou Eyecloud Technologies Co., Ltd.Inventors: Po Yuan, Shengjun Pan, Junneng Zhao, Daniel Mariniuc
-
Patent number: 10580025Abstract: The present disclosure describes systems and methods for automatically rolling-up data associated with one or more geographic units, such as ZIP+4 codes, such that the rollup comprises a minimum number of households to protect anonymity and ensure compliance with privacy regulations, while preserving variance of the underlying data associated with the geographic regions. Data attributes may include demographic data, socio-economic data, lifestyle segmentation, psychographic data, behavioral data, credit data, and other data. The rollup process may involve identifying one or more geographic units with a number of households below a minimum or threshold amount, applying filters to find candidate geographic units for rollup, scoring candidate geographic units to select best pairings for rollup, and repeating until the rollup group has at least the minimum number of households. The process may make trades off between granularity (e.g.Type: GrantFiled: September 27, 2018Date of Patent: March 3, 2020Assignee: Experian Information Solutions, Inc.Inventors: Andrew John Hickman, Alan Tsang, Yaqi Tao, Shengjun Pan
-
Publication number: 20200005613Abstract: A video surveillance method includes the steps of detecting, by a motion detector, an object motion, in the field of view of a surveillance device when the surveillance device is in a standby mode; generating, responsive to a positive detection to switch the surveillance device from the standby mode to an operation mode, one or more images of the moving object; determining, by processing the one or more images with a deep neural network (DNN) model of an object detector, whether the objects contained in the one or more images belong to a given categories, wherein the DNN model comprises N (N is a positive integer and ranged from 4-12) depthwise separable convolution layers; and video recording, responsive to a positive determination, the moving object in the field of view of the surveillance device.Type: ApplicationFiled: January 2, 2019Publication date: January 2, 2020Inventors: Po YUAN, Shengjun PAN, Junneng ZHAO, Daniel MARINIUC
-
Publication number: 20200005573Abstract: A smart door lock system provides an unlock authority of an electronically-controlled door lock mounted on a door to a remote computing device, thereby allowing the owner to remotely unlock the electronically-controlled door lock via the computing device rather than being physically present to perform the security check of the electronically-controlled door lock to open the door. Moreover, automatic transmission of the image data of the moving object in the field of view of a camera system in response to determining that one or more criteria are satisfied, facilitates door surveillance to help ensure personal and property's premise.Type: ApplicationFiled: January 2, 2019Publication date: January 2, 2020Inventors: Po YUAN, Shengjun PAN, Junneng ZHAO, Daniel MARINIUC
-
Publication number: 20190340904Abstract: A door surveillance system is adapted for implementing remote interaction between a visiting object and an owner of a property's premise and monitoring the area proximate to the door remotely. The door surveillance system comprises an interaction interface configured to receive an interaction request operation. Upon detecting an interaction request of the visiting object, at least a portion of the image data of the visiting object is outputted for transmission to the remote computing device along with the interaction request, thereby enabling the visiting object to interact with the owner of the property' premise. Automatic transmission of the image data of the visiting object facilitates door surveillance to help ensure personal and property's premise.Type: ApplicationFiled: July 3, 2019Publication date: November 7, 2019Inventors: Po Yuan, Shengjun Pan, Junneng Zhao, Daniel Mariniuc
-
Patent number: 10467313Abstract: To maximize the accuracy and efficiency of predicting users that will enjoy targeted content, a proposed content selection solution looks to combine a first strategy of utilizing selection rules with a second strategy of utilizing machine based learning models. By combining the selection rules-based approach and the machine learning model-based approach, the proposed content selection solution is able to consider and recommend a wider range of users for each available content.Type: GrantFiled: March 15, 2017Date of Patent: November 5, 2019Assignee: Oath Inc.Inventors: Liang Wang, Shengjun Pan, Kuang-chih Lee, Quan Lu, Junwei Pan
-
Publication number: 20190188593Abstract: The present teaching generally relates to removing perturbations from predictive scoring. In one embodiment, data representing a plurality of events detected by a content provider may be received, the data indicating a time that a corresponding event occurred and whether the corresponding event was fraudulent. First category data may be generated by grouping each event into one of a number of categories, each category being associated with a range of times. A first measure of risk for each category may be determined, where the first measure of risk indicates a likelihood that a future event occurring at a future time is fraudulent. Second category data may be generated by processing the first category data and a second measure of risk for each category may be determined. Measure data representing the second measure of risk for each category and the range of times associated with that category may be stored.Type: ApplicationFiled: December 19, 2017Publication date: June 20, 2019Inventors: Liang Wang, Angus Xianen Qiu, Shengjun Pan
-
Publication number: 20190095939Abstract: The present disclosure describes systems and methods for automatically rolling-up data associated with one or more geographic units, such as ZIP+4 codes, such that the rollup comprises a minimum number of households to protect anonymity and ensure compliance with privacy regulations, while preserving variance of the underlying data associated with the geographic regions. Data attributes may include demographic data, socio-economic data, lifestyle segmentation, psychographic data, behavioral data, credit data, and other data. The rollup process may involve identifying one or more geographic units with a number of households below a minimum or threshold amount, applying filters to find candidate geographic units for rollup, scoring candidate geographic units to select best pairings for rollup, and repeating until the rollup group has at least the minimum number of households. The process may make trades off between granularity (e.g.Type: ApplicationFiled: September 27, 2018Publication date: March 28, 2019Inventors: Andrew John Hickman, Alan Tsang, Yaqi Tao, Shengjun Pan
-
Patent number: 10102536Abstract: The present disclosure describes systems and methods for automatically rolling-up data associated with one or more geographic units, such as ZIP+4 codes, such that the rollup comprises a minimum number of households to protect anonymity and ensure compliance with privacy regulations, while preserving variance of the underlying data associated with the geographic regions. Data attributes may include demographic data, socio-economic data, lifestyle segmentation, psychographic data, behavioral data, credit data, and other data. The rollup process may involve identifying one or more geographic units with a number of households below a minimum or threshold amount, applying filters to find candidate geographic units for rollup, scoring candidate geographic units to select best pairings for rollup, and repeating until the rollup group has at least the minimum number of households. The process may make trades off between granularity (e.g.Type: GrantFiled: April 3, 2014Date of Patent: October 16, 2018Assignee: Experian Information Solutions, Inc.Inventors: Andrew John Hickman, Alan Tsang, Yaqi Tao, Shengjun Pan
-
Publication number: 20180268073Abstract: To maximize the accuracy and efficiency of predicting users that will enjoy targeted content, a proposed content selection solution looks to combine a first strategy of utilizing selection rules with a second strategy of utilizing machine based learning models. By combining the selection rules-based approach and the machine learning model-based approach, the proposed content selection solution is able to consider and recommend a wider range of users for each available content.Type: ApplicationFiled: March 15, 2017Publication date: September 20, 2018Applicant: Yahoo Holdings, Inc.Inventors: Liang Wang, Shengjun Pan, Kuang-chih Lee, Quan Lu, Junwei Pan