-
公开(公告)号：US11947570B2
公开(公告)日：2024-04-02
申请号：US16558869
申请日：2019-09-03
Applicant: International Business Machines Corporation
Inventor： Qing Wang , Shi Lei Zhang , Yonghua Lin
CPC classification number: G06F16/285 , G06N3/08
Abstract: A computer-implemented method for data augmentation is provided according an embodiment of the present disclosure. In the method, a first feature vector for input data may be obtained based on a first model. The input data may be clustered to a plurality of clusters. For each of the clusters, a second feature vector may be obtained based on the first model. Then, a similarity between the first feature vector and the second feature vector may be estimated for each of the clusters. At least one cluster of the plurality of clusters for which the similarity is lower than a threshold may be determined. Moreover, data augmentation may be performed to the at least one cluster.
-
公开(公告)号：US20220004759A1
公开(公告)日：2022-01-06
申请号：US16918598
申请日：2020-07-01
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Junsong Wang , Jie Zhang , Qing Wang , Yonghua Lin
Abstract: According to one embodiment, a method, computer system, and computer program product for generating anchors and selecting feature maps for a multi-scale object detection program based on analysis of the dataset is provided. The present invention may include generating a scale distribution of one or more scales of ground-truth objects, and, based on the scale distribution, dividing the effective scale range into a number of anchors greater than zero; furthermore, the invention may include generating a ratio distribution of ratios of the ground-truth objects; based on the ratio distribution, generating a ratio for at least one of the number of anchors; determining a template scale of one or more feature maps; and assigning the number of anchors to the feature maps based on the relative size of a scale of an anchor matching the relative size of a template scale of a feature map.
-
公开(公告)号：US20200167639A1
公开(公告)日：2020-05-28
申请号：US16199823
申请日：2018-11-26
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor： Chao Xue , Rong Yan , Yonghua Lin , Yonggang Hu , Yu Song
Abstract: Methods and systems for training a machine learning model include training a machine learning model using training data. A status of the machine learning model's training is determined based on an accuracy curve of the machine learning model over the course of the training. Parameters of the training are adjusted based on the status. Training of the machine learning model is completed using the adjusted parameters.
-
公开(公告)号：US10386445B2
公开(公告)日：2019-08-20
申请号：US15661642
申请日：2017-07-27
Applicant: International Business Machines Corporation
Inventor： YuBo Li , Yonghua Lin , Qing Wang , Wei Dong Wang , Chao Xue
Abstract: A method comprises receiving a first received signal strength indicator (RSSI) of a first beacon in an array of beacons and receiving a second RSSI of a second beacon in an array of beacons, calculating a RSSI of the array (r) as a function of the first RSSI and the second RSSI, retrieving a calibrated RSSI value of the array (r′) from a memory, determining whether r>r′, and outputting a signal to a user device responsive to determining that r>r′.
-
公开(公告)号：US09746541B2
公开(公告)日：2017-08-29
申请号：US14952036
申请日：2015-11-25
Applicant: International Business Machines Corporation
Inventor： YuBo Li , Yonghua Lin , Qing Wang , Wei Dong Wang , Chao Xue
Abstract: A method comprises receiving a first received signal strength indicator (RSSI) of a first beacon in an array of beacons and receiving a second RSSI of a second beacon in an array of beacons, calculating a RSSI of the array (r) as a function of the first RSSI and the second RSSI, retrieving a calibrated RSSI value of the array (r′) from a memory, determining whether r>r′, and outputting a signal to a user device responsive to determining that r>r′.
-
Patent Agency Ranking
公开(公告)号：US09641283B2
公开(公告)日：2017-05-02
申请号：US14597708
申请日：2015-01-15
Applicant: International Business Machines Corporation
Inventor： Yonghua Lin , Jianbin Tang , Jun Song Wang , Qi Jun Wang
CPC classification number: H04L1/0015 , H04L1/0003 , H04L1/0009 , H04L1/0026 , H04L1/0035 , H04L1/20
Abstract: An adaptive modulation coding method includes obtaining an estimated signal-to-interference-plus-noise ratio (SINR) and a packet error rate (PER) of a communication link of a user terminal; setting a first adjustment amount of the SINR of the communication link according to the PER of the communication link; calculating an effective SINR of the communication link according to the estimated SINR of the communication link and the first adjustment amount of the SINR of the communication link; and determining a modulation coding scheme of the communication link according to the effective SINR of the communication link.
-
公开(公告)号：US09264511B2
公开(公告)日：2016-02-16
申请号：US14830323
申请日：2015-08-19
Applicant: International Business Machines Corporation
Inventor： Li Li , Yonghua Lin , Rui Xiong Tian , Qing Wang , Yi Xin Zhao
CPC classification number: H04L67/2857 , H04L61/1582 , H04L67/1063 , H04L67/2804 , H04W4/023 , H04W4/21 , H04W24/10 , H04W28/16
Abstract: A method for realizing content sharing among mobile terminals comprises: receiving an index of access content sent by a requesting mobile terminal; looking up a shareable content index table for the index of the access content; and sending identification information of a mobile terminal having the index of the access content to the requesting mobile terminal if the index of the access content is found, so that the requesting mobile terminal obtains the requested access content from the mobile terminal having the index of the access content via wireless communication.
-
公开(公告)号：US20140016579A1
公开(公告)日：2014-01-16
申请号：US13928503
申请日：2013-06-27
Applicant: International Business Machines Corporation
Inventor： Xinhao Cheng , Yonghua Lin , Hang Liu , Jianbin Tang , Jun Song Wang , Qing Wang , Xian Wei Yi , Zhen Bo Zhu
IPC: H04W72/04
CPC classification number: H04W72/0453 , H04J11/0026 , H04L27/2647 , H04W72/1231
Abstract: This invention relates to wireless signal process, and discloses a method for interference process of an interference signal in an OFDM based wireless communication system, comprising: receiving an interference signal; performing time domain and frequency domain analyses on the interference signal to obtain interference characteristics of the interference signal; determining an interference category of the interference signal according to the interference characteristics of the interference signal; and performing interference process on the interference signal in view of the interference category of the interference signal. The method for interference signal process of this invention is effective to both homogeneous network interference and heterogeneous network interference, so that frequency spectrum sharing of homogeneous networks and existing heterogeneous networks can be realized.
Abstract translation: 本发明涉及无线信号处理，并公开了一种在基于OFDM的无线通信系统中干扰信号的干扰处理方法，包括：接收干扰信号; 对干扰信号进行时域和频域分析，获得干扰信号的干扰特性; 根据干扰信号的干扰特性确定干扰信号的干扰类别; 考虑到干扰信号的干扰类别，对干扰信号进行干扰处理。 本发明的干扰信号处理方法对于均匀网络干扰和异构网络干扰都是有效的，从而可以实现均匀网络和现有异构网络的频谱共享。
-
公开(公告)号：US20130321182A1
公开(公告)日：2013-12-05
申请号：US13925275
申请日：2013-06-24
Applicant: International Business Machines Corporation
Inventor： Xinhao Cheng , Mingqiang Li , Yonghua Lin , Qing Wang , Chao Xue , Rong Yan , Hai Zhan , Chao Zhu
IPC: H03M7/30
Abstract: The present invention relates to data compression and decompression, and particularly relates to a method and an apparatus for compressing and decompressing signal data. In an embodiment of the present invention, there is disclosed a method for compressing signal data, comprising: obtaining signal data; determining block lengths of a plurality of data blocks into which the signal data are divided, and determining exponents of the data blocks; forming a mantissa sequence of the data block by using the exponent of the data block to compress signal data contained in the data block; and forming a compressed data block by using the block length, the exponent and the mantissa sequence of the data block. By constructing a variable-length data block adapted to dynamic characteristics of signal data, the method for compressing signal data of the present invention increases the compression ratio of signal data.
-
公开(公告)号：US11373407B2
公开(公告)日：2022-06-28
申请号：US16663767
申请日：2019-10-25
Applicant: International Business Machines Corporation
Inventor： Qing Wang , Shi Lei Zhang , Yonghua Lin
Abstract: A computer-implemented method for attention generation is provided. In this method, a plurality of image frames can be obtained from a video stream. An original attention for a first image frame of the plurality of image frames can be generated. Then, at least one interested area can be identified in the first image frame. A local attention for each of the at least one interested area can be generated. Moreover, a total attention for the first image frame can be generated based on the original attention of the first image frame and the local attention of each of the at least one interested area.
-
-
-
-
-
-
-
-
-
Search Results
59
records
(took 0.022 seconds)
