公开(公告)号：US20180246925A1

公开(公告)日：2018-08-30

申请号：US15758072

申请日：2015-09-09

Applicant: Intel Corporation

Inventor： Guang-He Lee ,  Shao-Wen Yang

IPC: G06F17/30

CPC classification number: G06F16/2365 ,  G06F16/215 ,  G06F16/24568 ,  G06F17/10

Abstract: A set of data is identified that includes a plurality of observed values generated by a plurality of sensor devices located in a plurality of different locations. For each of the plurality of observed values, a modality of the value, a spatial location of the value, and a timestamp of the value is determined. Values for one or more missing values in the set of data are determined from the modalities, spatial locations, and timestamps of the plurality of observed values.

公开(公告)号：US10909335B2

公开(公告)日：2021-02-02

申请号：US16062105

申请日：2015-12-22

Applicant: Intel Corporation

Inventor： Guang-He Lee ,  Kalpana A. Algotar ,  Shao-Wen Yang ,  Addicam V. Sanjay

IPC: G06K7/10 ,  G01S1/04 ,  G01S13/75 ,  G06K9/62

Abstract: A set of samples are returned by radio frequency identifier (RFID) reader corresponding to the readings of signals emitted from a particular RFID tag, each sample including a respective set of features identifying values of the attributes of the signals as detected. At least some of the features are provided as inputs to a random forest of decision trees, each providing a prediction that the particular RFID tag is located in one of a plurality of defined zones in a particular environment. From outputs of the plurality of decision trees based on the set of samples, it can be determined that the particular RFID tag is located in a particular one of the plurality of zones at a first instance in time.

公开(公告)号：US20180375743A1

公开(公告)日：2018-12-27

申请号：US16062107

申请日：2015-12-26

Applicant: Guang-He LEE ,  Shao-Wen YANG ,  Intel Corporation

Inventor： Guang-He Lee ,  Shao-Wen Yang

IPC: H04L12/26 ,  H04L29/08 ,  G06F17/16 ,  G06K9/62

Abstract: A plurality of sensor data instances from a sensor device are identified and one or more tensors for a data set based on the plurality of sensor data instances is determined. A predicted value for each instance in the data set based on the tensors, as well as a predicted variance for each instance in the data set. A sampling rate to be applied at the sensor device is determined based on the predicted variances.

公开(公告)号：US20180373906A1

公开(公告)日：2018-12-27

申请号：US16062105

申请日：2015-12-22

Applicant: Intel Corporation

Inventor： Guang-He Lee ,  Kalpana A. Algotar ,  Shao-Wen Yang ,  Addicam V. Sanjay

IPC: G06K7/10 ,  G01S1/04 ,  G01S13/75 ,  G06K9/62

Abstract: A set of samples are returned by radio frequency identifier (RFID) reader corresponding to the readings of signals emitted from a particular RFID tag, each sample including a respective set of features identifying values of the attributes of the signals as detected. At least some of the features are provided as inputs to a random forest of decision trees, each providing a prediction that the particular RFID tag is located in one of a plurality of defined zones in a particular environment. From outputs of the plurality of decision trees based on the set of samples, it can be determined that the particular RFID tag is located in a particular one of the plurality of zones at a first instance in time.

公开(公告)号：US20210294788A1

公开(公告)日：2021-09-23

申请号：US16949531

申请日：2020-11-02

Applicant: Intel Corporation

Inventor： Guang-He Lee ,  Shao-Wen Yang

IPC: G06F16/23 ,  G06F16/2455 ,  G06F16/215

Abstract: A set of data is identified that includes a plurality of observed values generated by a plurality of sensor devices located in a plurality of different locations. For each of the plurality of observed values, a modality of the value, a spatial location of the value, and a timestamp of the value is determined. Values for one or more missing values in the set of data are determined from the modalities, spatial locations, and timestamps of the plurality of observed values.

公开(公告)号：US10824618B2

公开(公告)日：2020-11-03

申请号：US15758072

申请日：2015-09-09

Applicant: Intel Corporation

Inventor： Guang-He Lee ,  Shao-Wen Yang

IPC: G06F16/00 ,  G06F16/23 ,  G06F16/2455 ,  G06F16/215 ,  G06F17/10

Abstract: A set of data is identified that includes a plurality of observed values generated by a plurality of sensor devices located in a plurality of different locations. For each of the plurality of observed values, a modality of the value, a spatial location of the value, and a timestamp of the value is determined. Values for one or more missing values in the set of data are determined from the modalities, spatial locations, and timestamps of the plurality of observed values.

公开(公告)号：US20170068888A1

公开(公告)日：2017-03-09

申请号：US14757959

申请日：2015-12-24

Applicant: INTEL CORPORATION

Inventor： Yu-An Chung ,  Guang-He Lee ,  Shao-Wen Yang

IPC: G06N3/08 ,  G06N3/04

CPC classification number: G06N3/084 ,  G06N3/0454

Abstract: Classification techniques are disclosed that take into account the “cost” of each type of classification error for minimizing total cost of errors. In one example embodiment, a pre-trained cost-sensitive auto-encoder can be used in combination with a training (fine-tuning) stage for cost-sensitive deep learning. Thus, cost information is effectively combined with deep learning by modifying the objective function in the pre-training phase. By minimizing the modified objective function, the auto-encoder not only tries to capture underlying pattern, it further “learns” the cost information and “stores” it in the structure. By later fine-tuning at the training stage, the classification system yields improved performance (lower cost) than a typical classification system that does not take cost information into account during pre-training.

Abstract translation: 公开了分类技术，其考虑到每种类型的分类错误的“成本”，以最小化错误的总成本。 在一个示例实施例中，预训练的成本敏感型自动编码器可以与用于成本敏感的深度学习的训练（微调）阶段结合使用。 因此，通过修改训练前阶段的目标函数，成本信息与深度学习有效结合。 通过最小化修改的目标函数，自动编码器不仅尝试捕获底层模式，而且进一步“学习”成本信息并将其存储在结构中。 通过在培训阶段稍后进行微调，分类系统比预培训期间不考虑成本信息的典型分类系统产生改进的性能（降低成本）。