Microsoft Technology Licensing, LLC

International Business Machines Corporation

Journey Technology, Ltd.

Jingdong City (beijing) Digits Technology Co.,ltd.

Google Inc.

Lenovo (beijing) Limited

Beijing Bytedance Network Technology Co., Ltd.

Lenovo Limited

Yu Zheng

Xing Xie

Yu Chen Zhou

Xiao Bi Liang

Longhao Wang

Ning Zhang

Yin Lou

Chengyang Zhang

Like Liu

Wei-Ying Ma

Sharad Agarwal

Shaw-Ben Shi

Ting Li

Wayne B Riley

Meng Hong Tsai

Cheng Luo

Eric Chang

Junbo Zhang

Qingyan Wang

Chang Sheng

Alex Hehmeyer

Wuwen Ding

Zaiben Chen

Jing Chang Huang

Yi Ge

Chao Xue

Wei Jiang

Zhiqiang He

Li Song

Bo Zheng

Wei Zhu

Leonid Karlinsky

Hongtao Ma

Dan Wu

Chang Xu

Jiang Qian

Wen Xiong

Noor-E-Gagan Singh

Yuming Liang

Jiangwei Zhong

Xuan Li

Ruochi Zhang

Jian Jin

Weizhen Wang

Xingxiu Chen

Jianye Ye

Feng Ren

Menghua Zhang

Junkai Sun

Yanqing Xiao

Yang Ye

Xinling Wu

Min Lu

Mingdong Zhao

Hepeng Gao

Zhao Tai Pan

Nicholas Jing Yuan

Jianqiao Feng

Wencheng Zheng

Lizhu Zhang

Xinyan Han

Vincent Zhou

Venky Venkateshaiah

Jianguang Wu

Jun Zhi

Zhiqian Ma

Yongrong Wang

Xinsheng Huang

Cheng Wei

Using a training portion of a dataset, a set of component parameters comprising parameters of a component of an object detection model are trained. Using the trained set of component parameters, a set of backbone component weights comprising weights of component types in a backbone portion of the object detection model are trained. Using the trained set of component parameters, a set of backbone link weights comprising weights of links within the backbone portion are trained. Using the trained set of component parameters, a set of head component weights comprising weights of component types in a head portion of the object detection model are trained. Using the trained sets of component parameters, backbone component weights, backbone link weights, and head component weights, a trained object detection model is configured and trained to perform object detection.

Automated end-to-end machine learning model optimization

This disclosure describes a map-matching module that supports a Global Positioning System (GPS) and provides a user with a best match trajectory corresponding to GPS sampling points taken at a low sampling rate. The best match trajectory is based upon a spatial-temporal analysis.

Map-matching for low-sampling-rate GPS trajectories

An anomaly detection method for time series data includes: obtaining a plurality of pieces of feature information of time series data; generating a target feature combination based on the plurality of pieces of feature information; and performing anomaly result detection based on the target feature combination.

Anomaly monitoring method and apparatus for timing data, electronic device, and storage medium

A method for determining a spatial relationship, includes: obtaining a calculation request for the spatial relationship, the calculation request including a polygonal region and a set of report points; determining a plurality of first coded values corresponding to the polygonal region; determining a plurality of second coded values corresponding to the set of report points and a group of report points corresponding to each of the plurality of second coded values; determining a second coded value matching each of the plurality of first coded values by sorting and matching the plurality of first coded values and the plurality of second coded values respectively; and determining a relationship between the polygonal region and each report point in the set of report points according to a group of report points corresponding to the second coded value matching each of the plurality of first coded values.

Method and apparatus for determining spatial relationship, computer device, and storage medium

An optical apparatus includes an image source, a relay optical system, and an optical processing system. The image source is configured to display an image. The relay optical system is configured to project the image displayed by the image source to the optical processing system, and to image at infinity. The optical processing system is configured to project incident light from the relay optical system in a same direction to at least two preset directions sequentially.

Optical apparatus and waveguide display apparatus

According to embodiments of the present disclosure, there is provided a method, system, apparatus, electronic device, storage medium and program product for data exchange. The method described herein comprises: obtaining original data to be exchanged by a target application between a first platform and a second platform; obtaining normalized data corresponding to a type of the original data by processing the original data based on the type; determining a satisfaction of a data exchange constraint from the normalized data. In this way, the embodiments of the present disclosure may simplify and promote the determination of the data exchange constraint and accelerate the data exchange process.

Method, system, apparatus and device for data exchange

A system includes a laser emitting a laser beam, a beam expander expanding the laser beam, a beam splitter prism splitting the expanded laser beam into first and second split light beams, a liquid crystal box containing polymer-dispersed liquid crystal, first and second reflectors reflecting the first and second split light beams to the liquid crystal box, respectively, and an attenuator arranged on an optical path between the beam expander and the liquid crystal box. The attenuator gradually attenuates at least one of the laser beam, the expanded laser beam, the first split light beam, or the second split light beam along a first set curve. The first split light beam and the second split light beam form interference fringes at the liquid crystal box to expose the polymer-dispersed liquid crystal to form a polymer-dispersed liquid crystal holographic grating having a diffraction efficiency decreasing along a second set curve.

Preparation system and method for polymer-dispersed liquid crystal holographic grating

Facilitation of determination of detailed location of a source signal is provided. In one embodiment, a device comprises a memory that stores computer executable components; and a processor that executes computer executable components stored in the memory. The computer executable components can comprise: a low-resolution computation logic component that implements a coarse algorithm and determines an approximate direction of arrival (DOA) of a source signal of an input signal, wherein the coarse algorithm uses both a coarse spatial grid and input data received from the input signal to determine the approximate DOA; and an error estimation logic component that estimates an estimation error of the coarse algorithm, and wherein the error estimation logic component uses the estimation error and the approximate DOA to determine a spatial interval range.

Facilitation of efficient signal source location employing a coarse algorithm and high-resolution computation

A method, system, and computer program product for graph-based searching for one or more data streams is disclosed. A computer-implemented method comprises extracting a plurality of tuples from one or more data streams. The method further comprises generating a graph for the plurality of tuples in which a node represents a tuple of the plurality of tuples and an edge represents a correlation between the node and another node, and the edge is generated based at least partly on one or more predetermined queries for the one or more data streams. The method further comprises traversing the graph based on the one or more predetermined queries. Accordingly, embodiments of the present disclosure can improve the search speed by use of the graph-based searching for one or more data streams.

Graph-based searching for data stream

A computer implemented method, computer system and computer program product are provided for acoustic event detection in polyphonic acoustic data, according to the method, polyphonic acoustic data is inputted by one or more processing units into a trained neural network trained by labeled monophonic acoustic data, a first output from a hidden layer of the trained neural network is obtained by one or more processing units, and at least one acoustic classification of the polyphonic acoustic data is determined by one or more processing units based on the first output and a feature dictionary learnt from the trained neural network.

Acoustic event detection in polyphonic acoustic data

A method, computer system, and a computer program product for adaptively selecting an acoustic feature extractor in an Artificial Intelligence system is provided. The present invention may include acquiring a frame of an acoustic signal. The present invention may include checking a status of a flag to be used to indicate a proper acoustic feature extractor to be selected. The present invention may include processing the frame of the acoustic signal by the selected acoustic feature extractor indicated by the checked status. The present invention may include determining, based on data generated in the processing of the frame of the acoustic signal, an actual status of the frame of the acoustic signal. The present invention may include updating the status of the flag according to the actual status.

Acoustic feature extractor selected according to status flag of frame of acoustic signal

In various example embodiments, a system and method for generating and routing notifications of content extracted from emails are presented. A member's interest in a specific email among a set of emails are determined based on a relevance score. Content from the specific email are extracted, the content being identified as primary information and secondary information. A notification is generated comprising of the vital content extracted from the specific email. Presentation of the primary information is caused, the primary information being associated with the notification according to a ranked order, the ranked order being based on the relevance score. In response to a member action, presentation of the secondary information is caused, the secondary information being associated with the notification.

Generating and routing notifications of extracted email content

Growing community of inventors

Beijing, China

Yu Zheng