公开(公告)号：US20240111840A1

公开(公告)日：2024-04-04

申请号：US17957508

申请日：2022-09-30

申请人： Advanced Micro Devices, Inc. ,  ATI Technologies ULC

发明人： Akila Subramaniam ,  Ying Liu ,  Tung Chuen Kwong ,  Juanjo Noguera

IPC分类号： G06K9/62 ,  G06N3/04

CPC分类号： G06K9/6227 ,  G06K9/6261 ,  G06N3/04

摘要： An electronic device uses a tiling scheme selected from among a set of tiling schemes for processing instances of input data through a neural network. Each of the tiling schemes is associated with a different arrangement of portions into which instances of input data are divided for processing in the neural network. In operation, processing circuitry in the electronic device acquires information about a neural network and properties of the processing circuitry. The processing circuitry then selects a given tiling scheme from among a set of tiling schemes based on the information. The processing circuitry next processes instances of input data in the neural network using the given tiling scheme. Processing each instance of input data in the neural network includes dividing the instance of input data into portions based on the given tiling scheme, separately processing each of the portions in the neural network, and combining the respective outputs to generate an output for the instance of input data.

公开(公告)号：US20240086298A1

公开(公告)日：2024-03-14

申请号：US17940450

申请日：2022-09-08

申请人： Content Square SAS

发明人： Mohamed-Amine Baatout ,  Bharath Aasuri Sridharan ,  Daria Stefic

IPC分类号： G06F11/34 ,  G06F16/954 ,  G06F16/955 ,  G06K9/62 ,  G06N20/20

CPC分类号： G06F11/3438 ,  G06F16/954 ,  G06F16/955 ,  G06K9/6227 ,  G06N20/20

摘要： Systems and techniques may be used for generating an alert for a website based on user interactions at the website using a machine learning trained model. A technique may include receiving user interaction metrics corresponding to the user interactions, determining, using the machine learning trained model, whether a trajectory of a metric of the user interaction metrics is predicted to traverse a threshold, and in response to determining that the trajectory is predicted to traverse the threshold at a particular time, generating the alert. The technique may include evaluating a set of subsegments of the metric, each subsegment of the set of subsegments corresponding to respective metric source attributes, determining, based on the evaluation, at least one subsegment of the set of subsegments that most contributed to the trajectory, and outputting an indication for display, the indication including identification of the at least one subsegment with the alert.

公开(公告)号：US20230244991A1

公开(公告)日：2023-08-03

申请号：US17896281

申请日：2022-08-26

申请人： Databricks, Inc.

发明人： Benjamin Thomas Wilson ,  Corey Zumar

IPC分类号： G06N20/00 ,  G06K9/62

CPC分类号： G06N20/00 ,  G06K9/6227 ,  G06K9/6235 ,  G06K2009/6237

摘要： The present application discloses a method, system, and computer system for building a model associated with a dataset. The method includes receiving a data set, the dataset comprising a plurality of keys and a plurality of key-value relationships, determining a plurality of models to build based at least in part on the dataset, wherein determining the plurality of models to build comprises using the dataset format information to identify the plurality of models, building the plurality of models, and optimizing at least one of the plurality of models.

公开(公告)号：US20230237503A1

公开(公告)日：2023-07-27

申请号：US17648556

申请日：2022-01-21

申请人： Dell Products L.P.

发明人： Irfan Gilani ,  Sreenivas Sathyanarayana Sharma

IPC分类号： G06Q30/00 ,  G06Q30/02 ,  G06K9/62

CPC分类号： G06Q30/018 ,  G06Q30/0283 ,  G06K9/6227

摘要： In one aspect, an example methodology implementing the disclosed techniques includes, by an eco fees classification service, receiving information regarding a product to classify and generating a feature vector for the product, the feature vector representing a plurality of relevant features determined from the information regarding the product to classify. The method also includes, by the eco fee classification service, predicting, using an eco fees classification engine, a commodity classification for the product based on the feature vector, and recommending the commodity classification for the product for use in determining an eco fee to apply to a sale of the product. In some aspects, the method may also include computing the eco fee to apply to the sale of the product based on the recommended commodity classification.

公开(公告)号：US11645572B2

公开(公告)日：2023-05-09

申请号：US16831845

申请日：2020-03-27

申请人： NEC Laboratories Europe GmbH

发明人： Mischa Schmidt ,  Julia Gastinger

IPC分类号： G06F3/00 ,  G06K9/62 ,  G06F9/38 ,  G06N20/00 ,  G06F3/01

CPC分类号： G06K9/6227 ,  G06F9/3836 ,  G06K9/6262 ,  G06N20/00

摘要： A method for automatically selecting a machine learning algorithm and tuning hyperparameters of the machine learning algorithm includes receiving a dataset and a machine learning task from a user. Execution of a plurality of instantiations of different automated machine learning frameworks on the machine learning task are controlled each as a separate arm in consideration of available computational resources and time budget, whereby, during the execution by the separate arms, a plurality of machine learning models are trained and performance scores of the plurality of trained models are computed. One or more of the plurality of trained models are selected for the machine learning task based on the performance scores.

公开(公告)号：US20190220697A1

公开(公告)日：2019-07-18

申请号：US15870795

申请日：2018-01-12

申请人： MICROSOFT TECHNOLOGY LICENSING, LLC

发明人： Kenneth Liam KIEMELE ,  John Benjamin HESKETH ,  Evan Lewis JONES ,  James Lewis NANCE ,  LaSean Tee SMITH

IPC分类号： G06K9/62 ,  G06T7/73 ,  G06K9/00 ,  G06F15/18 ,  G06K9/32

CPC分类号： G06K9/6227 ,  G06K9/00335 ,  G06K9/00362 ,  G06K9/00771 ,  G06K9/3233 ,  G06K9/6262 ,  G06K9/6288 ,  G06N20/00 ,  G06N20/20 ,  G06T7/75 ,  H04W4/70

摘要： Techniques for generating a machine learning model to detect event instances from physical sensor data, including applying a first machine learning model to first sensor data from a first physical sensor at a location to detect an event instance, determining that a performance metric for use of the first machine learning model is not within an expected parameter, obtaining second sensor data from a second physical sensor during a period of time at the same location as the first physical sensor, obtaining third sensor data from the first physical sensor during the period of time, generating location-specific training data by selecting portions of the third sensor data based on training event instances detected using the second sensor data, training a second ML model using the location-specific training data, and applying the second ML model instead of the first ML model for detecting event instances.

公开(公告)号：US20180373964A1

公开(公告)日：2018-12-27

申请号：US15992246

申请日：2018-05-30

申请人： Hitachi, Ltd.

发明人： Yuki KONDO ,  Katsuto SATO

IPC分类号： G06K9/62 ,  G06T7/194 ,  G06K9/00

CPC分类号： G06K9/6268 ,  G06K9/00288 ,  G06K9/6227 ,  G06K9/6262 ,  G06K9/6274 ,  G06K9/6288 ,  G06K2209/27 ,  G06T5/50 ,  G06T7/11 ,  G06T7/194 ,  G06T2207/20081 ,  G06T2207/20221

摘要： Provided is an information processing apparatus configured to convert image data that has been input into saved data to save in a storage unit and reproduce the image data from the saved data. The information processing apparatus includes: an encoder unit configured to convert the image data into the saved data; and a decoder unit configured to reproduce the saved data as the image data. The encoder unit includes: a recognition unit configured to generate class tag information from the image data that has been input; a segmentation unit configured to generate region information that distinguishes a recognition target region and a background region from the image data that has been input; and a region separation unit configured to generate a background image according to the background region from the image data that has been input based on the region information.

公开(公告)号：US20180225125A1

公开(公告)日：2018-08-09

申请号：US15945276

申请日：2018-04-04

申请人： FUJITSU LIMITED

发明人： Hiroaki OKAMOTO ,  Tsuyoshi Nagato ,  Tetsuo Koezuka

IPC分类号： G06F9/44 ,  G06N3/12 ,  G06K9/62

CPC分类号： G06F9/44 ,  G06F8/36 ,  G06F8/44 ,  G06K9/00664 ,  G06K9/46 ,  G06K9/6227 ,  G06K9/6282 ,  G06N3/126

摘要： There is provided a program generating apparatus including a generating unit and a genetic processing unit. The generating unit is configured to generate tree structures each representing an image classification program. Each of the tree structures has a first level group and a second level group. Elements of nodes in the first level group are selected from amongst image filters each used to apply preprocessing to an input image. An element of a node in the second level group is selected from amongst setting programs each used to set a different value as a control parameter for generating a classifier based on information obtained by execution of the elements selected for the nodes in the first level group. The genetic processing unit is configured to output, using genetic programming, a tree structure with a fitness score exceeding a predetermined threshold based on the tree structures.

公开(公告)号：US20180218240A1

公开(公告)日：2018-08-02

申请号：US15459251

申请日：2017-03-15

申请人： Wipro Limited

发明人： Sendil Kumar Jaya KUMAR ,  Sujatha JAGANNATH

IPC分类号： G06K9/62 ,  G06K9/46

CPC分类号： G06K9/6267 ,  G06K9/00684 ,  G06K9/036 ,  G06K9/4604 ,  G06K9/4619 ,  G06K9/6227

摘要： The present disclosure is related in general to image processing and a method and system for generating a multi-level classifier for image processing. An image processing system may analyse an input image of a predetermined image type to extract unique key feature descriptors associated with the input image. Further, the unique key feature descriptors are resized into a predefined standard template format which is utilized to develop an image type classifier. Furthermore, the unique key feature descriptors are resized into each of one or more template classifiers of the predetermined image type. Further, signal quality value of each of the template classifiers is determined by validating each of the unique key feature descriptors resized based on each of the template classifiers and an image prediction classifier is developed based on the signal quality value.

公开(公告)号：US20180189605A1

公开(公告)日：2018-07-05

申请号：US15857032

申请日：2017-12-28

申请人： MorphoTrust USA, LLC

发明人： Yecheng Wu ,  Robert L. Jones ,  Brian K. Martin

IPC分类号： G06K9/62 ,  G07F7/08 ,  G07D7/005 ,  G07D7/2033 ,  B42D25/23 ,  G06K19/077 ,  G06K19/16 ,  G06K19/18 ,  H04N1/32

CPC分类号： G06K9/6227 ,  B42D25/23 ,  G06K9/00577 ,  G06K19/077 ,  G06K19/16 ,  G06K19/18 ,  G06K2009/0059 ,  G07D7/005 ,  G07D7/2008 ,  G07D7/2033 ,  G07F7/082 ,  G07F7/0826 ,  H04N1/32149

摘要： In some implementations, a system is capable of generating identifications that include distinctive line patterns corresponding to different portions of secure customer information. Data indicating an input image, and a dithering matrix representing a two-dimensional array of pixel values is obtained. Pixel values of pixels included in the input image are transformed using the dithering matrix. For each pixel within the input image, the transformation includes identifying a particular pixel value within the dithering matrix that represents a particular pixel within the input image, and adjusting an intensity value of the particular pixel based on attributes of the dithering matrix. A transformed image is generated based on the transformation and then provided for output.