Patents by Inventor Yongjun Chen

Yongjun Chen 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: 20240070744
    Abstract: Embodiments described herein provide systems and methods for training a sequential recommendation model. Methods include determining a difficulty and quality (DQ) score associated with user behavior sequences from a training dataset. User behavior sequences are sampled during training based on their DQ scores. A meta-extrapolator may also be trained based on user behavior sequences sampled according to DQ score. The meta-extrapolator may be trained with high quality low difficulty sequences. The meta-extrapolator may then be used with an input of high quality high difficulty sequences to generate synthetic user behavior sequences. The synthetic user behavior sequences may be used to augment the training dataset to fine-tune the sequential recommendation model, while continuing to sample user behavior sequences based on DQ score. As the DQ score is based on current model predictions, DQ scores iteratively update during the training process.
    Type: Application
    Filed: August 19, 2022
    Publication date: February 29, 2024
    Inventors: Yongjun Chen, Zhiwei Liu, Jianguo Zhang, Huan Wang, Caiming Xiong
  • Publication number: 20240062071
    Abstract: Embodiments described herein provide a cascade-guided adversarial training method for sequential recommendation models. A system may compute cascade scores for items in a user interaction sequence. The cascade scores may be based on the position in the sequence, as well as the appearance of the same item in other sequences. Based on the computed cascade score, the system may perturb item embeddings. The perturbed user interaction sequences with perturbed item embeddings may be used to train a sequential recommendation model.
    Type: Application
    Filed: December 30, 2022
    Publication date: February 22, 2024
    Inventors: Juntao TAN, Shelby Heinecke, Zhiwei Liu, Yongjun Chen
  • Publication number: 20240055807
    Abstract: An insert for a connector, said insert is provided having: an insert housing, a cavity of the insert housing having a plurality of connection accommodating holes for accommodating a plurality of connections, and the plurality of connections comprising at least one protective earthing connection; and a protective earthing plate which is attached to the outer surface of the insert housing in order to electrically contact the protective earthing connection. The insert can accommodate more connections, thereby providing a high-density connector; moreover, it is possible to use power connections and protective earthing connections of the same type in order to allow for quick assembly and disassembly, and customers can replace the connections according to actual use.
    Type: Application
    Filed: January 31, 2022
    Publication date: February 15, 2024
    Inventors: Yongjun CHEN, Yingtao WANG
  • Publication number: 20240046330
    Abstract: Embodiments described herein provide a universal item learning framework that generates universal item embeddings for zero-shot items. Specifically, the universal item learning framework performs generic features extraction of items and product knowledge characterization based on a product knowledge graph (PKG) to generate embeddings of input items. A pretrained language model (PLM) may be adopted to extract features from generic item side information, such as titles, descriptions, etc., of an item. A PKG may be constructed to represent recommendation-oriented knowledge, which comprise a plurality of nodes representing items and a plurality of edges connecting nodes represent different relations between items. As those relations in PKG are usually retrieved from user-item interactions, the PKG adapts the universal representation for recommendation with knowledge of user-item interactions.
    Type: Application
    Filed: March 13, 2023
    Publication date: February 8, 2024
    Inventors: Ziwei Fan, Yongjun Chen, Zhiwei Liu, Huan Wang
  • Patent number: 11783491
    Abstract: An object tracking method includes: obtaining a video stream acquired by a camera, and decoding the video stream to obtain a plurality of image frames; and obtaining position information of a target object in a current image frame. Obtaining the position information of the target object in the current image frame includes: performing, in response to that the current image frame is a first-type image, object detection on the current image frame to obtain the position information of the target object in the current image frame; and predicting, in response to that the current image frame is a second-type image, the position information of the target object in the current image frame based on motion vectors of a previous image frame and a position the target object in the previous image frame.
    Type: Grant
    Filed: September 10, 2021
    Date of Patent: October 10, 2023
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Mengyun Tang, Wen Zhou, Yongjun Chen
  • Publication number: 20230252345
    Abstract: Embodiments described herein provide methods and systems for training a sequential recommendation model. A system receives a plurality of user behavior sequences, and encodes those sequences into a plurality of user interest representations. The system predicts a next item using a sequential recommendation model, producing a probability distribution over a set of items. The next interacted item in a sequence is selected as a positive sample, and a negative sample is selected based on the generated probability distribution. The positive and negative samples are used to compute a contrastive loss and update the sequential recommendation model.
    Type: Application
    Filed: May 27, 2022
    Publication date: August 10, 2023
    Inventors: Yongjun Chen, Jia LI, Nitish Shirish Keskar, Caiming Xiong
  • Patent number: 11638696
    Abstract: The present invention provides a method for preparing an additive-free cell-wall-broken granule of a Chinese medicinal material. The cell-wall-broken granule is obtained by medicinal material concocting and grinding, first cell-wall-breaking and grinding, screening, second wall-breaking and grinding, mixing, pelleting, and granulating and sieving. The obtained granule has a moderate tightness, a relatively rapid dissolution and diffusion and a good uniformity, overcoming problems of a difficulty of dissolution and diffusion and a poor uniformity of the existing cell-wall-broken granules.
    Type: Grant
    Filed: October 21, 2020
    Date of Patent: May 2, 2023
    Assignee: ZHONGSHAN ZHONGZHI PHARMACEUTICAL GROUP CO., LTD.
    Inventors: Jinle Cheng, Zhitian Lai, Jiyin Xu, Yongjun Chen, Wen Deng, Lihua Peng, Weixuan Chen, Jinmei Chen, Yina Wang, Yanling Liang, Weilin Qiao, Xiaojun Cao
  • Patent number: 11605118
    Abstract: Embodiments described herein provide an attentive network framework that models dynamic attributes with item and feature interactions. Specifically, the attentive network framework first encodes basket item sequences and dynamic attribute sequences with time-aware padding and time/month encoding to capture the seasonal patterns (e.g. in app recommendation, outdoor activities apps are more suitable for summer time while indoor activity apps are better for winter). Then the attentive network framework applies time-level attention modules on basket items' sequences and dynamic user attributes' sequences to capture basket items to basket items and attributes to attributes temporal sequential patterns. After that, an intra-basket attentive module is used on items in each basket to capture the correlation information among items.
    Type: Grant
    Filed: December 4, 2020
    Date of Patent: March 14, 2023
    Assignee: salesforce.com, inc.
    Inventors: Yongjun Chen, Jia Li, Chenxi Li, Markus Anderle, Caiming Xiong, Simo Arajarvi, Harshavardhan Utharavalli
  • Publication number: 20230073754
    Abstract: Embodiments described herein provides an intent prototypical contrastive learning framework that leverages intent similarities between users with different behavior sequences. Specifically, user behavior sequences are encoded into a plurality of user interest representations. The user interest representations are clustered into a plurality of clusters based on mutual distances among the user interest representations in a representation space. Intention prototypes are determined based on centroids of the clusters. A set of augmented views for user behavior sequences are created and encoded into a set of view representations. A contrastive loss is determined based on the set of augmented views and the plurality of intention prototypes. Model parameters are updated based at least in part on the contrastive loss.
    Type: Application
    Filed: January 27, 2022
    Publication date: March 9, 2023
    Inventors: Yongjun Chen, Zhiwei Liu, Jia Li, Caiming Xiong
  • Publication number: 20230042327
    Abstract: A method for providing a neural network system includes performing contrastive learning to the neural network system to generate a trained neural network system. The performing the contrastive learning includes performing first model augmentation to a first encoder of the neural network system to generate a first embedding of a sample, performing second model augmentation to the first encoder to generate a second embedding of the sample, and optimizing the first encoder using a contrastive loss based on the first embedding and the second embedding. The trained neural network system is provided to perform a task.
    Type: Application
    Filed: January 19, 2022
    Publication date: February 9, 2023
    Inventors: Zhiwei Liu, Caiming Xiong, Jia Li, Yongjun Chen
  • Patent number: 11370708
    Abstract: A fiber cement material formulation comprising a cementitious binder, a siliceous material, fiber, alumina trihydrate and a bifunctional low density additive wherein the bifunctional low density additive comprises any one or more of diatomaceous earth, recycled autoclave fiber cement dust or cellulose dust. The fiber cement material formulation optionally further comprises a secondary low density additive which may be perlite. In some embodiments, a fiber cement article manufactured from the fiber cement material formulation comprises a density of approximately 1.1 g/cm3 or below.
    Type: Grant
    Filed: October 20, 2020
    Date of Patent: June 28, 2022
    Assignee: James Hardie Technology Limited
    Inventors: Xiangyuan Liu, Huagang Zhang, Jongmin Keum, Yongjun Chen, Yan Li, Weiling Peng, Thinh Pham, Farshad J. Motamedi, Lijun Mao, Thomas Mueller
  • Publication number: 20220115825
    Abstract: An adapter mechanism includes an electrical adapter and an attachment device. The electrical adapter includes an adapter housing and a pair of conductive prongs. The adapter housing includes a first complementary connector, and the pair of conductive prongs is configured to be inserted into an electrical outlet. The attachment device includes a fixation aperture and a second complementary connector. The attachment device is configured to move relative to the adapter housing between a retracted position and an extended position. When the attachment device is in the retracted position, the second complementary connector is uncoupled from the first complementary connector. When the attachment device is in the extended position, the fixation aperture is extended out from the adapter housing and the second complementary connector is coupled to the first complementary connector to secure the attachment device in place at the adapter housing.
    Type: Application
    Filed: October 12, 2020
    Publication date: April 14, 2022
    Inventors: Yao Wang, Yongjun Chen, Bin Shi
  • Patent number: 11285709
    Abstract: Disclosed herein are various embodiments of a protective packaging system suitable for use with a prefinished building article. The protective packaging system generally is designed to reduce shrink back of the flexible substrate after tension is applied. The protective packaging system generally comprises at least one outer layer, at least one core layer and at least one adhesive layer. One or more layers of the protective packaging system may include a roughening agent such that the flexible substrate will not “gloss” or burnish a coating on a prefinished building article. The protective packaging system may also include one or more biocides.
    Type: Grant
    Filed: May 15, 2020
    Date of Patent: March 29, 2022
    Assignee: James Hardie Technology Limited
    Inventors: Yongjun Chen, Caidian Luo, Jimmy Lee McCammon
  • Patent number: 11283224
    Abstract: An adapter mechanism includes an electrical adapter and an attachment device. The electrical adapter includes an adapter housing and a pair of conductive prongs. The adapter housing includes a first complementary connector, and the pair of conductive prongs is configured to be inserted into an electrical outlet. The attachment device includes a fixation aperture and a second complementary connector. The attachment device is configured to move relative to the adapter housing between a retracted position and an extended position. When the attachment device is in the retracted position, the second complementary connector is uncoupled from the first complementary connector. When the attachment device is in the extended position, the fixation aperture is extended out from the adapter housing and the second complementary connector is coupled to the first complementary connector to secure the attachment device in place at the adapter housing.
    Type: Grant
    Filed: October 12, 2020
    Date of Patent: March 22, 2022
    Assignee: Ademco Inc.
    Inventors: Yao Wang, Yongjun Chen, Bin Shi
  • Publication number: 20220062056
    Abstract: A protection cup (100) for a hearing protection device (HPD) and a method for manufacturing a protection cup (100). The protection cup (100) includes a plurality of fractal elements disposed on an inner surface (101) of the protection cup (100).
    Type: Application
    Filed: February 20, 2019
    Publication date: March 3, 2022
    Inventors: John JENKINS, Scott HUTT, Wei SUN, Yongjun CHEN, Lei JIANG
  • Publication number: 20220058714
    Abstract: Embodiments described herein provide an attentive network framework that models dynamic attributes with item and feature interactions. Specifically, the attentive network framework first encodes basket item sequences and dynamic attribute sequences with time-aware padding and time/month encoding to capture the seasonal patterns (e.g. in app recommendation, outdoor activities apps are more suitable for summer time while indoor activity apps are better for winter). Then the attentive network framework applies time-level attention modules on basket items' sequences and dynamic user attributes' sequences to capture basket items to basket items and attributes to attributes temporal sequential patterns. After that, an intra-basket attentive module is used on items in each basket to capture the correlation information among items.
    Type: Application
    Filed: December 4, 2020
    Publication date: February 24, 2022
    Inventors: Yongjun Chen, Jia Li, Chenxi Li, Markus Anderle, Caiming Xiong, Simo Arajarvi, Harshavardhan Utharavalli
  • Publication number: 20220033997
    Abstract: Disclosed are a polyurethane elastic fiber with a flame retardant function and a preparation method thereof. The polyurethane elastic fiber is prepared by using a polyether diol containing phosphorus elements or a polyester diol containing phosphorus elements as a raw material to react with 4,4?-diphenylmethane diisocyanate to prepare a prepolymer, extending the chain using an organic amine to obtain a polyurethane solution, and dry spinning with the polymer solution to prepare the polyurethane fiber. The limit oxygen index of the prepared polyurethane fiber was between 25% and 32%.
    Type: Application
    Filed: October 19, 2021
    Publication date: February 3, 2022
    Inventors: Yongjun CHEN, Qianli MA
  • Patent number: D942089
    Type: Grant
    Filed: February 20, 2019
    Date of Patent: January 25, 2022
    Assignee: HONEYWELL INTERNATIONAL INC.
    Inventors: John Jenkins, Scott Hutt, Wei Sun, Yongjun Chen, Lei Jiang
  • Patent number: D945538
    Type: Grant
    Filed: May 14, 2019
    Date of Patent: March 8, 2022
    Assignee: Zhongshan Shengdongli Fitness Equipment Co., Ltd.
    Inventor: Yongjun Chen
  • Patent number: D960459
    Type: Grant
    Filed: November 22, 2021
    Date of Patent: August 9, 2022
    Assignee: HONEYWELL INTERNATIONAL INC.
    Inventors: John Jenkins, Scott Hutt, Wei Sun, Yongjun Chen, Lei Jiang