International Business Machines Corporation

Empire Technology Development LLC

Samsung Electronics Co., Ltd.

Boe Technology Group Co., Ltd.

Lenovo (beijing) Limited

The Penn State Research Foundation

Chinese Academy of Sciences

Beijing Lenovo Software Ltd.

Other

Tsinghua University

Beijing Boe Display Technology Co., Ltd.

Baidu Online Network Technology (beijing) Co., Ltd.

Rensselaer Polytechnic Institute

Beijing Boe Technology Development Co., Ltd.

Beijing Bytedance Network Technology Co., Ltd.

Qi Cheng Li

Li Jun Mei

Jian Wang

Xin Zhou

Shao Chun Li

Hao Chen

Ya Bin Dang

Xuefeng Song

Jie Ma

Lijun Mei

Feng Li

Zi Yu Zhu

Fang Quan Xie

Hong Wang

Jonathan D Dunne

Lixiang Xu

Xiaowan Ke

Juhnyoung Lee

Rong Liu

Guoqiang Hu

Wei Sun

Xin Zhang

Bo Yang

Xiao Ling Chen

Shuai Chen

Chuang Gan

Bo Tong Liu

Guo Qiang Hu

Xiuzhu Tang

Wei Ye Chen

Rangachari Anand

Zachary A Silverstein

Zhen Liu

Kelley Anders

Zhi Zhang

Cheng Fang Wang

David Nahamoo

Zhihua Sun

Ke Xu

Guang Yang

Bo Wu

Hui Li

Soumitra Sarkar

Leho Nigul

Yan Zhao

Tian Xia

Yinlong Zhang

Shulin Yao

Hailong Wu

Sijia Liu

Xuebo Liang

Yong He

Lei Zhang

Liam S Harpur

Wei Liu

Yi Wang

Yingyang Li

Fan Yang

Trudy L Hewitt

Ting Wang

Qian Wang

Yang Liu

Jing Li

Xibin Shao

Shichang Zhang

Wei He

Yanping Liao

Song Song

Huan Wang

Fan Jing Meng

Yi Bu Li

Dong Hui Liu

Qing Shan Zhang

Ajay Gupta

Fan Wang

Zhi Hong Li

Yan Gao

Da Li Liu

Shuang Hu

Jianfeng Yuan

Li Cao

Mary Swift

Yong Tian

Anindya Neogi

Chang Li

Jiang Yi Liu

Kyounghwan Choi

Shi Wan Zhao

Ke Shang

Jia Yu

Xiao Ping Guo

Yang Cao

Zhenguo Tian

Zhan Peng Huo

Guan Zhang

Lijun Xiong

Girish Bhimrao Chafle

Yong Yin

Sheng Lu

Evaluation of a health condition includes obtaining a correlation between dietary habits and health issues of a plurality of reference users. The correlation is determined based on reference patterns. Each of the reference patterns indicates temporal characteristics of at least one dietary habit and at least one health issue of a reference user of the plurality of reference users. A target pattern of a target user is determined at least in part based on a profile for the target user. The target pattern at least indicates temporal characteristics of at least one dietary habit of the target user. An evaluation of a health condition of the target user is generated based on the target pattern and the correlation.

Health condition evaluation based on temporal analysis of dietary habits

New question and answer (QA) pairs can be automatically discovered from a corpus of data such as online chats and conversations. Newly discovered QA pairs can augment QA database, which can be used by a computer processor or device, e.g., by a chatbot, an automated machine, and/or another. Existing QA knowledge can be used to learn the structures of QA knowledge distribution in conversations, and new QA knowledge can be automatically learned through the structure of learned QA knowledge distribution in conversations. The structure of learned QA knowledge distribution can be refined by adding more semantics based on labeled data.

Discovering new question and answer knowledge from conversation

An image rendering method and apparatus, a device, and a medium. The method includes: determining a pre-captured initial rendering video stream for a background image; obtaining initial rendering coordinates of each rendering pixel in each initial rendering image frame in the video stream, and obtaining a rendering moment of each initial rendering image frame in the initial rendering video stream; determining the amount of noise of each rendering pixel at the corresponding rendering moment according to the corresponding initial rendering coordinates, a preset noise texture, and the corresponding rendering moment, and obtaining target rendering coordinates of each rendering pixel according to the corresponding amount of noise; and updating the initial rendering coordinates of each rendering pixel according to the corresponding target rendering coordinates to obtain a target rendering image, and rendering the background image according to a target rendering video stream composed of all target rendering images.

Image rendering method and apparatus, device, and medium

Disclosed is a disentangled personalized federated learning method via consensus representation extraction and diversity propagation provided by embodiments of the present application. The method includes: receiving, by a current node, local consensus representation extraction models and unique representation extraction models corresponding to other nodes, respectively; extracting, by the current node, the representations of the data of the current node by using the unique representation extraction models of other nodes respectively, and calculating first mutual information between different sets of representation distributions, determining similarity of the data distributions between the nodes based on the size of the first mutual information, and determining aggregation weights corresponding to the other nodes based on the first mutual information; the current node obtains the global consensus representation aggregation model corresponding to the current node.

Disentangled personalized federated learning method via consensus representation extraction and diversity propagation

Disclosed is a method for updating a node model that resists discrimination propagation in federated learning. The method includes: obtaining a node model corresponding to a data node; calculating a mean value of the distribution of class features and a quantity ratio corresponding to training data of the data node, calculating a distribution weighted aggregation model based on the node model, the mean value of the distribution of class features and the quantity ratio; calculating a regularization term corresponding to the data node based on the node model and the distribution weighted aggregation model; calculating a variance of the distribution of the class features corresponding to the data node, calculating a class balanced complementary term by using a cross-domain feature generator; and updating the node model based on the distribution weighted aggregation model, the regularization term, and the class balanced complementary term.

Method for updating a node model that resists discrimination propagation in federated learning

A dynamic path verification method based on reorganization of authentication fragments is proposed. The method includes: sending an initial expected path verification structure to a data packet sending end via a guarantee service node, and sending notification information to the respective routing nodes on an initial expected path; after updating the initial expected path to a new expected path, inserting the new expected path verification structure into a subsequent data packet to be sent, verifying the data packet by other nodes except a migration node, and sending the subsequent data packet to be sent to a next hop of routing node; performing the parsing verification on the received data packet by the migration node.

Dynamic path verification method based on reorganization of authentication fragments

Using data of a plurality of interactions within a collaborative network, a collaborative network graph is constructed. Each node in the collaborative network graph represents a participant in an interaction, and each edge in the collaborative network graph represents an interaction between participants represented by corresponding nodes. Using a plurality of stochastic simulations of changes to the collaborative network graph, integration of a new node into the collaborative network is simulated. A simulation in the plurality of stochastic simulations producing a largest score improvement between scores computed on the collaborative network graph is selected as a best simulation. Using a plurality of changes to the collaborative network graph included in the best simulation, an integration recommendation is constructed, in which a portion of the integration recommendation corresponds to a change in the plurality of changes.

Dynamically integrating nodes into a collaborate network

The disclosure provides a collaborative deep learning method and a collaborative deep learning apparatus. The method includes: sending an instruction for downloading a global model to a plurality of user terminals; receiving a set of changes from each user terminal; storing the set of changes; recording a hash value of the set of changes into a blockchain; obtaining a storage transaction number from the blockchain for the hash value of the set of changes; sending the set of changes and the storage transaction number to the plurality of user terminals; receiving the set of target user terminals from the blockchain; updating the current parameters of the global model based on sets of changes corresponding to the set of target user terminals; and returning the sending the instruction, to update the global model until the global model meets a preset condition.

Collaborative deep learning methods and collaborative deep learning apparatuses

A method and an apparatus of establishing dual-connectivity to transmit data are disclosed. A PCell of the UE carries out a mapping function and realizes a PDCP layer, a RLC layer, a MAC layer and a physical layer, and a SCell of the UE carries out quality packet data to data radio bearer mapping and realizes a PDCP layer, a MAC layer and a physical layer, in which a core network transmits data of the UE to the PCell, the PCell maps a QoS Flow of the data of the UE which is quality packet data to a data radio bearer on a mapping layer, performs path split, transmits data of respective paths to the UE through the PCell and SCell of the UE, and the UE regroups the data of the respective paths on the PDCP layer, and transmits the regrouped data to an application layer.

Method and apparatus for establishing dual-connectivity to transmit data in new radio communication architecture

The present disclosure provides a chip-on-film, a display substrate, a display device and a driving method. The display substrate includes a plurality of scanning lines and a plurality of cascaded shift register units coupled to the scanning lines. The display substrate further includes an output control signal line electrically coupled to each shift register unit for providing an enable signal for the shift register unit. The output control signal lines include at least two output control signal lines for providing different enable signals, and two adjacent cascaded shift register units are coupled to different output control signal lines.

Chip-on-film, display substrate, display device, and driving method

A communication method and system for controlling and paging a user equipment (UE) are provided. The method includes receiving mobility level information of a UE, and performing at least one of handover control, measurement control, or paging for the UE according to mobility level information of the UE.

Method and apparatus for controlling or paging a user equipment

Aspects include determining a coding intention and a dimension of interest to a user. A plurality of relevant projects that each include a logical code block that meets the coding intention are located. The locating includes searching a plurality of code repositories based at least in part on the coding intention. A score is assigned to each of the plurality of logical code blocks based at least in part on properties associated with the logical code blocks and on the dimension of interest to the user. A logical code block with the highest score is promoted to the user.

Dynamic code snippet promotion

A video generation method includes: obtaining a target face image and a source face image; extracting a feature of each of the source face image and the target face image through a face feature encoder, to obtain corresponding source feature codes and target feature codes; generating swapped face feature codes through a face feature exchanger according to the source feature codes and the target feature codes; generating an initial swapped face image through a face generator according to the swapped face feature codes; and fusing the initial swapped face image with the target face image through a face fuser, to obtain a final swapped face image. The face feature encoder performs hierarchical encoding on the face feature to reserve semantic details of a face, and the face feature exchanger performs further processing based on the hierarchical encoding, to obtain hierarchical encoding of a swapped face feature with semantic details.

Video generation method and system for high resolution face swapping

Systems, methods and/or computer program products for automating evaluations of code deliveries for software projects. Automated evaluations are generated by selecting software projects and dividing a project into a plurality of code blocks by analyzing the code, build file and build log. Attributes of code blocks are extracted and correlations between code blocks are calculated. Dynamic distance between code blocks is calculated by the correlation and code delivery history, to create a Dynamic General Distance Map and Dynamic User Distance Map for the code blocks. Code delivery distance indicating the code delivery behavior is generated by the Dynamic User Distance Map while the delivery assessment criteria is generated to evaluate code delivery and assess levels of risk associated with delivery of the code. High-level risk indicates that a code delivery may not follow best practices and users are alerted to pay more attention to the code delivery during review.

Automated evaluation of code delivery

A level conversion circuit and a display panel are provided. The level conversion circuit includes a signal generation circuit configured to output driving signals through a plurality of signal output terminals, a first operational amplification circuit configured to level-convert a voltage of an input terminal and output the voltage through an output terminal, and signal output terminals of the signal generation circuit are in one-to-one correspondence with input terminals of the first operational amplification circuit; a plurality of switching circuits connected between the signal output terminal and the input terminal, connected to a control signal terminal, and configured to communicate the signal output terminal with the input terminal in response to a signal of the control signal terminal. At least part of the switching circuits are connected to different control signal terminals.

Level conversion circuit, and display panel

Growing community of inventors

Beijing, China

Qi Cheng Li