Patents by Inventor Zhe Zhao
Zhe Zhao 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: 12265455Abstract: The present invention relates to a method, system and computer program product for task failover in an unstable environment, wherein the unstable environment includes a plurality of reclaimable nodes. According to the method, it is monitored if any node of the plurality of reclaimable nodes is to be reclaimed. Whether a task on any node of the plurality of reclaimable nodes is recoverable is determined. Responsive to the task being recoverable, data of the recoverable task is stored. Responsive to a node being reclaimed and the task on the reclaimed node being recoverable, at least one associated task of at least one associated node of the reclaimed node is notified to wait.Type: GrantFiled: October 29, 2021Date of Patent: April 1, 2025Assignee: International Business Machines CorporationInventors: Guang Han Sui, Wei Ge, Lan Zhe Liu, Zhang Li Ping, Er Tao Zhao
-
Patent number: 12259937Abstract: Embodiments relate to graph-based design mapping and function route generation. A computer-implemented method includes analyzing first user interfaces and second user interfaces to identify features of each user interface, where the second user interfaces include at least one change to the first user interfaces. A first graph representing the features of the first user interfaces and a second graph representing the features of the second user interfaces are generated. A function mapping is determined based on a similarity calculation between the first graph and the second graph. A node that satisfies an influence threshold is recommended to do limited manual labeling. The function mapping is updated based on the labeling. A visualization of the function mapping is created between the first user interfaces and the second user interfaces. Function route generation can be performed.Type: GrantFiled: March 6, 2023Date of Patent: March 25, 2025Assignee: International Business Machines CorporationInventors: Rong Zhao, Zhe Yan, Li Li Guan, Hao Jia Li, Li Bo Zhang
-
Publication number: 20250095328Abstract: Method and apparatus for video processing. A plurality of videos are accessed. An index is generated based on one or more key frames and one or more video features extracted from the plurality of video. A user input is processed to identify user intentions. The index is filtered to identify a plurality of candidate key frames based on the user intentions. A recommended video is generated based on the plurality of candidate key frames, further comprising selecting one or more additional frames based on processing the plurality of candidate key frames using one or more clustering operations, and aggregating the plurality of candidate key frames and the one or more additional frames.Type: ApplicationFiled: September 19, 2023Publication date: March 20, 2025Inventors: Li Bo ZHANG, Zhe YAN, Rong ZHAO, Li Li GUAN, Hao Xiang WU
-
Publication number: 20250089638Abstract: According to one embodiment, a method, computer system, and computer program product for pollen prediction is provided. The present invention may include identifying, by a classification model, plant species within a plurality of locatable images taken at a plurality of locations; creating growth cycle prediction models for the plant species; modelling pollen count mappings for the plant species; predicting pollen yields at the locations for the plant species based on the growth cycle prediction models and the pollen count mappings; and calculating a pollen distribution at the locations based on the predicted pollen yields and aerodynamic models.Type: ApplicationFiled: September 19, 2023Publication date: March 20, 2025Inventors: Li Li Guan, Zhe Yan, Rong Zhao, Li Bo Zhang, Hao Xiang Wu
-
Publication number: 20250085115Abstract: A computer-implemented method of trajectory prediction includes obtaining a first cross-attention between a vectorized representation of a road map near a vehicle and information obtained from a rasterized representation of an environment near the vehicle by processing through a first cross-attention stage; obtaining a second cross-attention between a vectorized representation of a vehicle history and information obtained from the rasterized representation by processing through a second cross-attention stage; operating a scene encoder on the first cross-attention and the second cross-attention; operating a trajectory decoder on an output of the scene encoder; obtaining one or more trajectory predictions by performing one or more queries on the trajectory decoder.Type: ApplicationFiled: November 3, 2023Publication date: March 13, 2025Inventors: Hao XIAO, Yiqian GAN, Ethan ZHANG, Xin YE, Yizhe ZHAO, Zhe HUANG, Lingting GE, Robert August ROSSI, JR.
-
Patent number: 12248457Abstract: An embodiment for detecting anomalous data using dependency modeling. The embodiment may, within a target data environment, identify references between data contained in one or more data files. The embodiment may determine dependency relationships between data fields in the data contained in the one or more data files. The embodiment may construct computational graphs depicting the determined dependency relationships as series of related data fields. The embodiment may identify a series of associated computational graphs within the constructed computational graphs. The embodiment may calculate abnormality degree values for each of the data fields within the constructed computation graphs. The embodiment may, in response to detecting an anomalous data field having a calculated abnormality degree value above a threshold value, calculating contribution values for a series of associated component data fields to identify a root cause for the detected anomalous data field.Type: GrantFiled: August 23, 2023Date of Patent: March 11, 2025Assignee: International Business Machines CorporationInventors: Hao Xiang Wu, Rong Zhao, Zhe Yan, Li Li Guan, Li Bo Zhang
-
Publication number: 20250074463Abstract: A method of predicting vehicle trajectory includes operating a scene encoder on an environmental representation surrounding a vehicle; concatenating an output of the scene encoder with a history trajectory; applying a sequence encoder to a result of the concatenating; refining an output of the sequence encoder based on the history trajectory; and generating one or more predicted future trajectories by operating a decoder on an output of the refining.Type: ApplicationFiled: February 6, 2024Publication date: March 6, 2025Inventors: Ethan ZHANG, Hao XIAO, Yiqian GAN, Yizhe ZHAO, Zhe HUANG, Lingting GE
-
Publication number: 20250068619Abstract: An embodiment for detecting anomalous data using dependency modeling. The embodiment may, within a target data environment, identify references between data contained in one or more data files. The embodiment may determine dependency relationships between data fields in the data contained in the one or more data files. The embodiment may construct computational graphs depicting the determined dependency relationships as series of related data fields. The embodiment may identify a series of associated computational graphs within the constructed computational graphs. The embodiment may calculate abnormality degree values for each of the data fields within the constructed computation graphs. The embodiment may, in response to detecting an anomalous data field having a calculated abnormality degree value above a threshold value, calculating contribution values for a series of associated component data fields to identify a root cause for the detected anomalous data field.Type: ApplicationFiled: August 23, 2023Publication date: February 27, 2025Inventors: Hao Xiang Wu, Rong Zhao, Zhe Yan, Li Li Guan, Li Bo Zhang
-
Patent number: 12235316Abstract: Chip moving device is provided. The chip moving device includes a base, including a mounting portion that includes an end surface and an assembly cavity extending in a horizontal direction and penetrating through the end surface, the end surface being arranged with a limiting component; an adsorption part, extending in a vertical direction and confined by the limiting component; and a clamping component, arranged in the assembly cavity and including an end extending out of the end surface of the mounting portion, the clamping component being configured to be telescopic in the horizontal direction, and provide the adsorption part with a clamping force in the horizontal direction when the clamping component is extended or contracted.Type: GrantFiled: January 30, 2024Date of Patent: February 25, 2025Assignee: SEMIGHT INSTRUMENTS CO., LTDInventors: Zhe Lian, Jianjun Huang, Yonghong Wu, Shan Zhao, Haiyang Hu
-
Publication number: 20250054286Abstract: An image processing method includes performing, using images obtained from one or more sensors onboard a vehicle, a 2-dimensional (2D) feature extraction; performing, a 3-dimensional (3D) feature extraction on the images; detecting objects in the images by fusing detection results from the 2D feature extraction and the 3D feature extraction.Type: ApplicationFiled: September 27, 2023Publication date: February 13, 2025Inventors: Zhe HUANG, Lingting GE, Yizhe ZHAO
-
Publication number: 20250050913Abstract: Techniques are described for operating a vehicle using sensor data provided by one or more ultrasonic sensors located on or in the vehicle. An example method includes receiving, by a computer located in a vehicle, data from an ultrasonic sensor located on the vehicle, where the data includes a first set of coordinates of two points associated with a location where an object is detected by the ultrasonic sensor; determining a second set of coordinates associated with a point in between the two points; performing a first determination that the second set of coordinates is associated with a lane or a road on which the vehicle is operating; performing a second determination that the object is movable; and sending, in response to the first determination and the second determination, a message that causes the vehicle to perform a driving related operation while the vehicle is operating on the road.Type: ApplicationFiled: October 13, 2023Publication date: February 13, 2025Inventors: Zhe CHEN, Lingting GE, Joshua Miguel RODRIGUEZ, Ji HAN, Panqu WANG, Junjun XIN, Xiaoling HAN, Yizhe ZHAO
-
Publication number: 20250051104Abstract: The present disclosure includes a die inspection sorting system and a formation method. The die inspection sorting system includes a turntable and a plurality of die picking assemblies installed at an equal interval along a circumferential direction at an edge of the turntable. The die picking assembly includes a vertical support frame, a base installed on a front side of the vertical support frame, a suction rod installed vertically on a side of the base away from the vertical support frame, and a clamping strip. The limiting groove is disposed on a front end surface of the base; an installation groove is disposed on the front end surface of the base; a movable block is inserted in the installation groove; and a spring in a compressed state is connected between a rear end of the movable block and a bottom surface of the installation groove.Type: ApplicationFiled: February 5, 2024Publication date: February 13, 2025Inventors: Renwei TANG, Zhe LIAN, Yonghong WU, Shan ZHAO
-
Publication number: 20250042369Abstract: Techniques are described for determining a set of pose information for an object when multiple sets of pose information are determined for a same object from multiple images. An example driving operation method includes obtaining, by a computer located in a vehicle, at least two sets of pose information related to an object located on a road on which the vehicle is operating, where each set of pose information includes characteristic(s) about the object, and where each set of pose information is determined from an image obtained by a camera; determining at least two weighted output vectors; determining, for the object, a set of pose information that are based on a combined weighted output vector that is obtained by combining the at least two weighted output vectors; and causing the vehicle to perform a driving-related operation using the set of pose information for the object.Type: ApplicationFiled: October 13, 2023Publication date: February 6, 2025Inventors: Yizhe ZHAO, Zhe CHEN, Lingting GE, Panqu WANG
-
Patent number: 12216139Abstract: The present disclosure includes an adaptive die test socket and a formation method. The die test socket includes an upper test socket and a lower test socket disposed directly below the upper test socket. The upper test socket is connected to a support frame; a strip-shaped through hole is formed on an upper surface of the support frame; a first protruding strip is at each of lower portions of two opposite inner walls of the strip-shaped through hole; a corresponding second protruding strip is at each of upper portions of two opposite side surfaces of the installation plate; and a side of a strip-shaped block is fixedly connected to the upper surface of the support frame, and another side of the strip-shaped block extends to directly above the second protruding strip and is connected to the second protruding strip through at least two springs.Type: GrantFiled: February 5, 2024Date of Patent: February 4, 2025Assignee: SEMIGHT INSTRUMENTS CO., LTDInventors: Zhe Lian, Jianjun Huang, Yonghong Wu, Shan Zhao, Haiyang Hu
-
Patent number: 12216156Abstract: The present disclosure includes a die test pressing-down apparatus and a formation method. The die test pressing-down apparatus includes a support frame capable of moving along a vertical direction; and a pressing-down block installed on the support frame. A part to-be-pressed is disposed directly below the pressing-down block. The pressing-down block is connected to the support frame through an installation plate; a strip-shaped through hole is formed at an upper surface of the support frame; a first protruding strip is at each lower portion of two opposite inner walls of the strip-shaped through hole; a corresponding second protruding strip is at each upper portion of two opposite side surfaces of the installation plate; and a side of a strip-shaped block is fixedly connected to the upper surface of the support frame, and another side of the strip-shaped block is connected to the second protruding strip through at least two springs.Type: GrantFiled: February 5, 2024Date of Patent: February 4, 2025Assignee: SEMIGHT INSTRUMENTS CO., LTDInventors: Zhe Lian, Jianjun Huang, Yonghong Wu, Shan Zhao, Haiyang Hu
-
Patent number: 12216907Abstract: The present disclosure provides a method for performing a programming operation on a memory cell connected to a bit line and controlled by a word line. The method includes applying a first programming voltage signal to the word line to program the memory cell into a first state, applying a first voltage to the bit line, performing a verify operation when the memory cell is in a second state, determining a classification of the memory cell based on the verify operation, applying a second voltage to the bit line based on the determined classification, applying a second programming voltage signal to the word line to program the memory cell into the first state, applying a third voltage to the bit line, applying a third programming voltage signal to the word line to program the memory cell into the first state, and applying a fourth voltage to the bit line.Type: GrantFiled: December 15, 2022Date of Patent: February 4, 2025Assignee: YANGTZE MEMORY TECHNOLOGIES CO., LTD.Inventors: Zhe Luo, Da Li, Feng Xu, Yaoyao Tian, Jianquan Jia, XiangNan Zhao
-
Publication number: 20250029274Abstract: The present disclosure provides methods and systems of sampling-based object pose determination. An example method includes obtaining, for a time frame, sensor data of the object acquired by a plurality of sensors; generating a two-dimensional bounding box of the object in a projection plane based on the sensor data of the time frame; generating a three-dimensional pose model of the object based on the sensor data of the time frame and a model reconstruction algorithm; generating, based on the sensor data, the pose model, and multiple sampling techniques, a plurality of pose hypotheses of the object corresponding to the time frame, generating a hypothesis projection of the object for each of the pose hypotheses by projecting the pose hypothesis onto the projection plane; determining evaluation results by comparing the hypothesis projections with the bounding box; and determining, based on the evaluation results, an object pose for the time frame.Type: ApplicationFiled: October 17, 2023Publication date: January 23, 2025Inventors: Yizhe ZHAO, Zhe CHEN, Ye FAN, Lingting GE, Zhe HUANG, Panqu WANG, Xue MEI
-
Publication number: 20250020687Abstract: The present disclosure provides an adaptive die testing apparatus and a formation method. The adaptive die testing apparatus includes a support frame capable of moving along a vertical direction; a movable panel capable of moving along a horizontal direction; an upper test socket installed on the support frame; and a lower test socket installed on the movable panel and capable of moving with the movable plate to directly below the upper test socket. A groove is formed on a lower surface of the upper test socket and/or an upper surface of the lower test socket. When the adaptive die testing apparatus is at a testing state, the lower surface of the upper test socket is in a close contact with the upper surface of the lower test socket, and a sealed chamber for placing a die to-be-tested is formed at a region of the groove.Type: ApplicationFiled: February 6, 2024Publication date: January 16, 2025Inventors: Zhe LIAN, Jianjun HUANG, Yonghong WU, Shan ZHAO, Haiyang HU
-
Publication number: 20240386280Abstract: A computer-implemented method to generate a second machine learning model based on a first machine learning model, wherein the second machine learning model is structured for more efficient computation, is provided. The method includes processing an input with a hidden layer of a student machine-learned model to obtain an intermediate output. The method includes providing an encoded message descriptive of the input and the intermediate output for processing with a teacher machine-learned model. The method includes, responsive to providing the encoded message, obtaining a second encoded message descriptive of a second intermediate output of one or more hidden layers of the teacher machine-learned model. The method includes performing a knowledge distillation training process to train the student machine-learned model based on a difference between the intermediate output and the second intermediate output.Type: ApplicationFiled: May 17, 2024Publication date: November 21, 2024Inventors: Zhe Zhao, Huan Gui, Qingyun Liu, Ed Huai-hsin Chi, Lichan Hong, Bang An
-
Publication number: 20240388362Abstract: Aspects of this technical solution are directed to multiplexed transmission by optical beam transmission. A method can include generating a first beam having a first frequency in an optical frequency range, generating a second beam including data and having a second frequency in the optical frequency range, converting the first beam and the second beam into a third beam including the data and having a third frequency based on a difference between the first frequency and the second frequency, and transforming the third beam into a fourth beam having a first angular distribution and including a first portion of the data, and into a fifth beam having a second angular distribution corresponding to the first angular distribution and including a second portion of the data.Type: ApplicationFiled: September 13, 2022Publication date: November 21, 2024Inventors: Alan Willner, Amir Minoofar, Xinzhou Su, Zhe Zhao, Huibin Zhou, Runzhou Zhang