18th week of 2020 patent applcation highlights part 57 |
Patent application number | Title | Published |
20200134339 | Image Processing for Identifying Individuals - Cameras capture time-stamped images of predefined areas. At least one image includes a representation of a portion of an individual with other portions of the individual occluded within the image. Pixel attributes for the portion of the individual are identified and provided as a box or set of coordinates for tracking the individual within the image and in subsequent images taken. | 2020-04-30 |
20200134340 | STARTUP AUTHENTICATION METHOD FOR INTELLIGENT TERMINAL - Disclosed is a startup authentication method for an intelligent terminal, including first performing face authentication, and continuing to perform gesture-based virtual password authentication after the face authentication, wherein even if the face authentication is cracked, the gesture-based password authentication is required to perform for logging in, so the disclosure can effectively improve the security of authentication. Further, in the disclosure, the gesture-based virtual password authentication is performed based on a gesture image input by a user in the air, so that since there is no need to perform input operations on a screen of the intelligent terminal, the aesthetics of the intelligent terminal will not be affected. Moreover, in the disclosure, when the virtual password is determined by detecting binary images of fingertips, the disturbance of the binary images of the fingertips is also removed, which can improve the probability and efficiency in subsequent detection of the virtual password. | 2020-04-30 |
20200134341 | INTELLIGENT TERMINAL - Disclosed is an intelligent terminal, for which the startup authentication includes first performing face authentication and continuing to perform gesture-based virtual password authentication after the face authentication, even if the face authentication is cracked, the gesture-based password authentication is required to perform for logging in, and so the intelligent terminal of the disclosure can effectively improve the security of authentication. Further, the gesture-based virtual password authentication is performed based on a gesture image input by a user in the air, so that since there is no need to perform input operations on a screen of the intelligent terminal, the aesthetics of the intelligent terminal will not be affected. Moreover, in the disclosure, when the virtual password is determined by detecting binary images of fingertips, the disturbance of the binary images of the fingertips is also removed, which can improve the probability and efficiency in subsequent detection of the virtual password. | 2020-04-30 |
20200134342 | SPOOF DETECTION USING MULTIPLE IMAGE ACQUISITION DEVICES - The technology described in this document can be embodied in a method that includes receiving from a sensor, information indicative of an environmental condition. The method also includes receiving first information indicative of whether or not a first image captured by a first image acquisition device corresponds to an alternative representation of a live person, and receiving second information indicative of whether or not a second image captured by a second image acquisition device corresponds to the alternative representation. The first information and the second information are combined in a weighted combination, the corresponding weights being assigned in accordance with the environmental condition. A determination is made, based on the weighted combination, that a subject in the first and second images is an alternative representation of a live person, and in response, access to the secure system is prevented. | 2020-04-30 |
20200134343 | COLLATION DEVICE AND COLLATION METHOD - A collation device is configured to include a processor, and a storage unit that stores a predetermined determination condition in advance, under which a photographic image which is an image obtained by imaging a photograph of the subject is capable of being eliminated, the processor is configured to detect brightness distribution of a face image obtained by imaging an authenticated person with an imaging unit, determine whether or not the detected brightness distribution satisfies a determination condition, and perform face authentication using the face image satisfying the determination condition. | 2020-04-30 |
20200134344 | SPOOF DETECTION USING STRUCTURED LIGHT ILLUMINATION - The technology described in this document can be embodied in a method that includes a method for preventing access to a secure system based on determining a captured image to be of an alternative representation of a live person. The method includes illuminating a subject with structured light using a light source array comprising multiple light sources disposed in a predetermined pattern, capturing an image of the subject as illuminated by the structured light, and determining that the image includes features representative of the predetermined pattern. The method also includes, responsive to determining that the image includes features representative of the predetermined pattern, identifying the subject in the image to be an alternative representation of a live person. The method further includes responsive to identifying the subject in the image to be an alternative representation of a live person, preventing access to the secure system. | 2020-04-30 |
20200134345 | SPOOF DETECTION USING IRIS IMAGES - The technology described in this document can be embodied in a method for preventing access to a secure system based on determining a captured image to be of an alternative representation of a live person. The method includes capturing an image of a subject illuminated by an infrared (IR) illumination source, and extracting, from the image, a portion representative of an iris of the subject. The method also includes determining that an amount of high-frequency features in the portion of the image satisfies a threshold condition indicative of the image being of an alternative representation of a live person, and in response, identifying the subject in the image to be an alternative representation of a live person. Responsive to identifying the subject in the image to be an alternative representation of a live person, the method further includes preventing access to the secure system. | 2020-04-30 |
20200134346 | INTELLIGENT GALLERY MANAGEMENT FOR BIOMETRICS - A system provides intelligent gallery management for biometrics. A first gallery is obtained that includes biometric and/or other information on a population of people. An application is identified. A subset of the population of people is identified based on the application. A second gallery is derived from the first gallery by pulling the information for the subset of the population of people without pulling the information for the population of people not in the subset. Biometric identification (such as facial recognition) for the application may then be performed using the second gallery rather than the first gallery. In this way, the system is improved as less time is required for biometric identification, fewer device resources are used, and so on. | 2020-04-30 |
20200134347 | INTELLIGENT GALLERY MANAGEMENT FOR BIOMETRICS - A system provides intelligent gallery management for biometrics. A first gallery is obtained that includes biometric and/or other information on a population of people. An application is identified. A subset of the population of people is identified based on the application. A second gallery is derived from the first gallery by pulling the information for the subset of the population of people without pulling the information for the population of people not in the subset. Biometric identification (such as facial recognition) for the application may then be performed using the second gallery rather than the first gallery. In this way, the system is improved as less time is required for biometric identification, fewer device resources are used, and so on. | 2020-04-30 |
20200134348 | OPTICAL CHARACTER RECOGNITIONS VIA CONSENSUS OF DATASETS - An example of apparatus includes a memory to store a first image of a document and a second image of the document. The first image and the second image are Memory captured under different conditions. The apparatus includes a processor coupled to the memory. The processor is to perform optical character recognition on the first image to generate a first output dataset and to perform optical character recognition on the second image to generate a second output dataset. The processor is further to determine whether consensus for a character is achieved based on a comparison of the first output dataset with the second output dataset, and generate a final output dataset based on the consensus for the character. | 2020-04-30 |
20200134349 | APPARATUS AND PROGRAM - A correction history recording unit that records region information of a correction site with respect to text data converted from an original image as correction history information, an accuracy calculation unit that calculates accuracy of optical character recognition for each of individual regions on a layout of the original image on the basis of the correction history information, a distribution image generation unit and a distribution image display unit which generate and display a distribution image in which a difference in magnitude of accuracy is shown as a difference in a display aspect for every individual region are included so as to generate and display the distribution image that is distinguished for every individual region by reflecting a tendency in which a character recognition rate in a certain region on a layout of the original image may decrease due to various cases including a format of an original document, a state of an OCR device, and the like. | 2020-04-30 |
20200134350 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND PROGRAM FOR DETECTING DEFECT FROM IMAGE - An image processing apparatus includes detecting means for detecting a first detected region and a second detected region from an input image, on the basis of a first detection criterion and a second detection criterion, respectively; image setting means for setting, as a target image subjected to correction, an image including the first detected region, and setting, as a reference image that is referred to in the correction, an image including the second detected region; accepting means for accepting, from a user, designation of a region in the target image and a correction instruction for the designated region; correction region setting means for identifying, in the reference image, a region corresponding to the designated region, and for setting a to-be-corrected region on the basis of the identified region and the second detected region; and correcting means for correcting the first detected region in the target image on the basis of the to-be-corrected region set in the reference image. | 2020-04-30 |
20200134351 | HANDHELD ARTHROPOD DETECTION DEVICE - Various embodiments include systems and methods of arthropod detection using an electronic arthropod detection device. The electronic arthropod detection device may scan a surface or a subject using a terahertz sensor that is sensitive to a terahertz band of electromagnetic radiation to detect the presence or likely presence of an arthropod in a region of interest (ROI). A camera sensitive to a visible band of electromagnetic radiation captures at least one image and provides the image(s) to an object detection model in response to determining that an arthropod is or is likely present in the ROI. A processor may initiate an arthropod detected procedure in response to detecting an arthropod in the ROI. | 2020-04-30 |
20200134352 | IMAGE ALIGNMENT METHOD AND DEVICE THEREFOR - Provided is a method for automatically performing image alignment without a user input. An image alignment method performed by an image alignment device, according to one embodiment of the present invention, can comprise the steps of: recognizing at least one person in an inputted image; determining a person-of-interest among the recognized persons; and performing image alignment, on the basis of the person-of-interest, on the inputted image, wherein the image alignment is performed without an input of a user of the image alignment device for the image alignment. | 2020-04-30 |
20200134353 | MONITORING-SCREEN-DATA GENERATION DEVICE, MONITORING-SCREEN-DATA GENERATION METHOD, AND REECORIDNG MEDIUM - A monitoring-screen-data generation device includes an object-data generation unit, a screen-data generation unit, and an assignment processing unit. The object-data generation unit identifies a plurality of objects included in an image based on image data, and generates object data. The screen-data generation unit generates monitoring screen data on the basis of the object data. On the basis of definition data that defines a state transition and the object data, the assignment processing unit assigns data that defines the state transition to an image object included in a monitoring screen of the monitoring screen data. | 2020-04-30 |
20200134354 | METHOD AND APPARATUS FOR PREDICTING FEATURE SPACE DECAY USING VARIATIONAL AUTO-ENCODER NETWORKS - An apparatus, method and computer program product are provided for predicting feature space decay using variational auto-encoder networks. Methods may include: receiving a first image of a road segment including a feature disposed along the road segment; applying a loss function to the feature of the first image; generating a revised image, where the revised image includes a weathered iteration of the feature; generating a predicted image using interpolation between the image and the revised image of a partially weathered iteration of the feature; receiving a user image, where the user image is received from a vehicle traveling along the road segment; correlating a feature in the user image to the partially weathered iteration of the feature in the predicted image; and establishing that the feature in the user image is the feature disposed along the road segment. | 2020-04-30 |
20200134355 | IMAGE PROCESSING SYSTEM AND COMPUTER PROGRAM FOR PERFORMING IMAGE PROCESSING - An object of the present invention is to achieve both suppression of data amount of an image processing system that learns a collation image to be used for image identification using a discriminator and improvement of identification performance of the discriminator. In order to achieve the above object, there is proposed an image processing system including a discriminator that identifies an image using a collation image, the image processing system further including a machine learning engine that performs machine learning of collation image data required for image identification. The machine learning engine searches for a successfully identified image using an image for which identification has been failed, and adds information, obtained based on a partial image of the image for which identification has been failed and which has been selected by an input device to the successfully identified image obtained by the search to generate corrected collation image data. | 2020-04-30 |
20200134356 | EFFICIENT SIMD IMPLEMENTATION OF 3X3 NON MAXIMA SUPPRESSION OF SPARSE 2D IMAGE FEATURE POINTS - In accordance with disclosed embodiments, an image processing method includes performing a first scan in a first direction on a first list of pixels in which, for each pixel in the first list, a feature point property is compared with a corresponding feature point property of each of a first set of neighboring pixels, performing a second scan in a second direction on the first list of pixels in which, for each pixel in the first list, a feature point property is compared with a corresponding feature point property of each of a second set of neighboring pixels, using the results of the first and second scans to identify pixels from the first list to be suppressed, and forming a second list of pixels that includes pixels from the first list that are not identified as pixels to be suppressed. The second list represents a non-maxima suppressed list. | 2020-04-30 |
20200134357 | NEURAL-NETWORK-BASED OPTICAL CHARACTER RECOGNITION USING SPECIALIZED CONFIDENCE FUNCTIONS - Systems and methods for neural-network-based optical character recognition using specialized confidence functions. An example method comprises: receiving a grapheme image; computing, by a neural network, a feature vector representing the grapheme image in a space of image features; and computing a confidence vector associated with the grapheme image, wherein each element of the confidence vector reflects a distance, in the space of image features, between the feature vector and a center of a class of a set of classes, wherein the class is identified by an index of the element of the confidence vector. | 2020-04-30 |
20200134358 | DETECTING INFECTION OF PLANT DISEASES WITH IMPROVED MACHINE LEARNING - A system and processing methods for refining a convolutional neural network (CNN) to capture characterizing features of different classes are disclosed. In some embodiments, the system is programmed to start with the filters in one of the last few convolutional layers of the initial CNN, which often correspond to more class-specific features, rank them to hone in on more relevant filters, and update the initial CNN by turning off the less relevant filters in that one convolutional layer. The result is often a more generalized CNN that is rid of certain filters that do not help characterize the classes. | 2020-04-30 |
20200134359 | CLUSTER VISUALIZATION DEVICE - A cluster visualization apparatus is disclosed. A cluster visualization apparatus according to the present disclosure includes a state detector configured to obtain state information of a cluster configured with a plurality of boxes, a display, and a controller configured to display a three-dimensional model image configured with a plurality of layers corresponding to a plurality of network layers and to display an image corresponding to each of the plurality of boxes over at least one layer of the plurality of layers, based on the state information. | 2020-04-30 |
20200134360 | Methods for Decreasing Computation Time Via Dimensionality - Dimensionality reduction in high-dimensional datasets can decrease computation time, and processes for dimensionality reduction may even be useful in lower-dimensional datasets. It has been discovered that methods of dimensionality reduction may dramatically decrease computational requirements in machine learning programming techniques. This development unlocks the ability of computational modeling to be used to solve complex problems that, in the past, would have required computation time on orders of magnitude too great to be useful. | 2020-04-30 |
20200134361 | DATA PROCESSING METHOD AND APPARATUS - The method includes: obtaining a plurality of pieces of feature data; automatically performing two different types of nonlinear combination processing operations on the plurality of pieces of feature data to obtain two groups of processed data, where the two groups of processed data include a group of higher-order data and a group of lower-order data, the higher-order data is related to a nonlinear combination of m pieces of feature data in the plurality of pieces of feature data, and the lower-order data is related to a nonlinear combination of n pieces of feature data in the plurality of pieces of feature data, where m≥3, and m>n≥2; and determining prediction data based on a plurality of pieces of target data, where the plurality of pieces of target data include the two groups of processed data. | 2020-04-30 |
20200134362 | SYSTEM AND METHOD OF CONNECTION INFORMATION REGULARIZATION, GRAPH FEATURE EXTRACTION AND GRAPH CLASSIFICATION BASED ON ADJACENCY MATRIX - Disclosed is system and method of connection information regularization, graph feature extraction and graph classification based on adjacency matrix. By concentrating the connection information elements in the adjacency matrix into a specific diagonal region of the adjacency matrix in order to reduce the non-connection information elements in advance. The subgraph structure of the graph is further extracted along the diagonal direction using the filter matrix. Then a stacked convolutional neural network is used to extract a larger subgraph structure. On the one hand, it greatly reduces the amount of computation and complexity, solving the limitations of the computational complexity and the limitations of window size. And on the other hand, it can capture large subgraph structure through a small window, as well as deep features from the implicit correlation structures at both vertex and edge level, which improves the accuracy and speed of the graph classification. | 2020-04-30 |
20200134363 | AUTOMATIC FEATURE SELECTION AND MODEL GENERATION FOR LINEAR MODELS - Methods, systems, and devices for automated feature selection and model generation are described. A device (e.g., a server, user device, database, etc.) may perform model generation for an underlying dataset and a specified outcome variable. The device may determine relevance measurements (e.g., stump R-squared values) for a set of identified features of the dataset and can reduce the set of features based on these relevance measurements (e.g., according to a double-box procedure). Using this reduced set of features, the device may perform a least absolute shrinkage and selection operator (LASSO) regression procedure to sort the features. The device may then determine a set of nested linear models—where each successive model of the set includes an additional feature of the sorted features—and may select a “best” linear model for model generation based on this set of models and a model quality criterion (e.g., an Akaike information criterion (AIC)). | 2020-04-30 |
20200134364 | Simultaneous Hyper Parameter and Feature Selection Optimization Using Evolutionary Boosting Machines - Aspects relate to a machine learning system implementing an evolutionary boosting machine. The system may initially select randomized feature sets for an initial generation of candidate models. Evolutionary algorithms may be applied to the system to create later generations of the cycle, combining and mutating the feature selections of the candidate models. The system may determine optimal number of boosting iterations for each candidate model in a generation by building boosting iterations from an initial value up to a predetermined maximum number of boosting iterations. When a final generation is achieved, the system may evaluate the optimal model of the generation. If the optimal boosting iterations of the optimal model does not meet solution constraints on the optimal boosting iterations, the system may adjust a learning rate parameter and then proceed to the next cycle. Based on termination criteria, the system may determine a resulting/final optimal mode. | 2020-04-30 |
20200134365 | INSTANCE SEGMENTATION METHODS AND APPARATUSES, ELECTRONIC DEVICES, PROGRAMS, AND MEDIA - An instance segmentation method includes: performing feature extraction on an image via a neural network to output features at at least two different hierarchies; extracting region features corresponding to at least one instance candidate region in the image from the features at the at least two different hierarchies, and fusing region features corresponding to a same instance candidate region, to obtain a first fusion feature of each instance candidate region; and performing instance segmentation based on each first fusion feature, to obtain at least one of an instance segmentation result of the corresponding instance candidate region or an instance segmentation result of the image. | 2020-04-30 |
20200134366 | TARGET RECOGNITION METHOD AND APPARATUS FOR A DEFORMED IMAGE - An object recognition method and apparatus for a deformed image are provided. The method includes: inputting an image into a preset localization network to obtain a plurality of localization parameters for the image, wherein the preset localization network comprises a preset number of convolutional layers, and wherein the plurality of localization parameters are obtained by regressing image features in a feature map that is generated from a convolution operation on the image; performing a spatial transformation on the image based on the plurality of localization parameters to obtain a corrected image; and inputting the corrected image into a preset recognition network to obtain an object classification result for the image. In the process of the neural network based object recognition, the embodiment of the present application first transforms the deformed image that has deformation, and then performs the object recognition on the transformed image. | 2020-04-30 |
20200134367 | INTERACTIVE MACHINE LEARNING MODEL DEVELOPMENT - A method is provided that includes generating a visual environment for interactive development of a machine learning (ML) model. The method includes accessing observations of data each of which includes values of independent variables and a dependent variable, and performing an interactive exploratory data analysis (EDA) of the values of a set of the independent variables. The method includes performing an interactive feature construction and selection based on the interactive EDA, and in which select independent variables are selected as or transformed into a set of features for use in building a ML model to predict the dependent variable. The method includes building the ML model using a ML algorithm, the set of features, and a training set produced from the set of features and observations of the data. And the method includes outputting the ML model for deployment to predict the dependent variable for additional observations of the data. | 2020-04-30 |
20200134368 | MACHINE LEARNING MODEL DEVELOPMENT WITH INTERACTIVE EXPLORATORY DATA ANALYSIS - A method is provided that includes generating a visual environment for interactive development of a machine learning (ML) model. The method includes accessing observations of data each of which includes values of independent variables and a dependent variable, and performing an interactive exploratory data analysis (EDA) of the values of a set of the independent variables. The method includes performing a feature construction and selection based on the interactive EDA, and in which select independent variables are selected as or transformed into a set of features for use in building a ML model to predict the dependent variable. The method includes building the ML model using a ML algorithm, the set of features, and a training set produced from the set of features and observations of the data. And the method includes outputting the ML model for deployment to predict the dependent variable for additional observations of the data. | 2020-04-30 |
20200134369 | MACHINE LEARNING MODEL DEVELOPMENT WITH INTERACTIVE FEATURE CONSTRUCTION AND SELECTION - A method is provided that includes generating a visual environment for interactive development of a machine learning (ML) model. The method includes accessing observations of data each of which includes values of independent variables and a dependent variable. The method includes performing an interactive feature construction and selection in which select independent variables are selected as or transformed into a set of features for use in building a ML model to predict the dependent variable. The method includes building the ML model using a ML algorithm, the set of features, and a training set produced from the set of features and observations of the data. And the method includes outputting the ML model for deployment to predict the dependent variable for additional observations of the data. | 2020-04-30 |
20200134370 | MACHINE LEARNING MODEL DEVELOPMENT WITH INTERACTIVE MODEL EVALUATION - A method is provided that includes generating a visual environment for interactive development of a machine learning (ML) model. The method includes accessing observations of data each of which includes values of independent variables and a dependent variable, and performing an interactive exploratory data analysis (EDA) of the values of a set of the independent variables. The method includes performing an interactive feature construction and selection based on the interactive EDA, and in which select independent variables are selected as or transformed into a set of features for use in building a ML model to predict the dependent variable. The method includes building the ML model using a ML algorithm, the set of features, and a training set produced from the set of features and observations of the data. And the method includes outputting the ML model for deployment to predict the dependent variable for additional observations of the data. | 2020-04-30 |
20200134371 | SYSTEMS AND METHODS FOR PROVIDING PERSONALIZED PRODUCT RECOMMENDATIONS USING DEEP LEARNING - Embodiments of the present disclosure provide systems, methods, and computer-readable media that use machine learning models to enable computing devices to detect and identify cosmetic products in face images. In some embodiments, a model training system may gather training data for building the machine learning models by analyzing face images associated with tagging data. In some embodiments, a recommendation system may be configured to use the machine learning models generated by the model training system to detect products in face images, and to add information based on the detected products to a look data store, and/or to provide recommendations for similar looks from the look data store based on the detected products. | 2020-04-30 |
20200134372 | METHODS AND SYSTEMS FOR THE FAST ESTIMATION OF THREE-DIMENSIONAL BOUNDING BOXES AND DRIVABLE SURFACES USING LIDAR POINT CLOUDS - The present invention relates to methods and systems for generating annotated data for training vehicular driver assist (DA) and autonomous driving (AD) active safety (AS) functionalities and the like. More specifically, the present invention relates to methods and systems for the fast estimation of three-dimensional (3-D) bounding boxes and drivable surfaces using LIDAR point clouds and the like. These methods and systems provide fast and accurate annotation cluster pre-proposals on a minimally-supervised or unsupervised basis, segment drivable surfaces/ground planes in a bird's-eye-view (BEV) construct, and provide fast and accurate annotation cluster pre-proposal labels based on the feature-based detection of similar objects in already-annotated frames. The methods and systems minimize the expertise, time, and expense associated with the manual annotation of LIDAR point clouds and the like in the generation of annotated data for training machine learning (ML) algorithms and the like. | 2020-04-30 |
20200134373 | MACHINE LEARNING DEVICE, DATA PROCESSING SYSTEM, PRINTING SYSTEM, MACHINE LEARNING METHOD, AND DATA PROCESSING METHOD - A machine learning device includes a state variable acquiring section, a teaching data acquiring section, and a learned model generating section. The state variable acquiring section acquires, as state variables: feature information that is information regarding a feature of an actual printed matter on which printing has been actually performed by an image forming apparatus; medium information that is information regarding a print medium used in the actual printed matter; and first control information that is information regarding control performed when the actual printed matter has been outputted. The teaching data acquiring section acquires, as teaching data, second control information that is information regarding control that causes the feature information to fall within a predetermined threshold. The learned model generating section generates a learned model by performing machine learning on the basis of the pieces of information acquired by the state variable acquiring section and the teaching data. | 2020-04-30 |
20200134374 | DYNAMIC ARTIFICIAL INTELLIGENCE / MACHINE LEARNING MODEL UPDATE, OR RETRAIN AND UPDATE, IN DIGITAL PROCESSES AT RUNTIME - Dynamically updating, or retraining and updating, artificial intelligence (AI)/machine learning (ML) models in digital processes at runtime is disclosed. Production operation may not need to be stopped for AI/ML model update or retraining and update. The update steps and/or retraining steps for the AWL model may be included as part of the digital process. The AI/ML model update may be requested from internal logic (e.g., from the evaluation of a condition, by an that expression calls for the AI/ML model, etc.), external requests (e.g., from external triggers in a finite state machine (FSM), such as a file change, database data, a service call, etc.), or both. Automation of AI/ML model updates or retraining and updates may be provided, where the software reloads/reinitializes/re-instantiates with a retrained and/or updated AWL model after (and possibly immediately after) the AI/ML model becomes available. | 2020-04-30 |
20200134375 | SEMANTIC SEGMENTATION MODEL TRAINING METHODS AND APPARATUSES, ELECTRONIC DEVICES, AND STORAGE MEDIA - A semantic segmentation model training method includes: performing, by a semantic segmentation model, image semantic segmentation on at least one unlabeled image to obtain a preliminary semantic segmentation result as the category of the unlabeled image; obtaining, by a convolutional neural network based on the category of the at least one unlabeled image and the category of at least one labeled image, sub-images respectively corresponding to the at least two images and features corresponding to the sub-images, where the at least two images comprise the at least one unlabeled image and the at least one labeled image, and the at least two sub-images carry the categories of the corresponding images; and training the semantic segmentation model on the basis of the categories of the at least two sub-images and feature distances between the at least two sub-images. | 2020-04-30 |
20200134376 | COMPUTER ARCHITECTURE FOR AND-OR NEURAL NETWORKS - A computer architecture for an and-or neural network is disclosed. A computing machine accesses an input vector. The input vector comprises a numeric representation of an input to a neural network. The computing machine provides the input vector to the neural network comprising a plurality of ordered layers. The plurality of ordered layers are alternating AND-layers and OR-layers. Each of the plurality of ordered layers receives input from a preceding layer and/or provides output to a next layer. The computing machine generates an output of the neural network based on an output of a last one of the plurality of ordered layers in the neural network. | 2020-04-30 |
20200134377 | LOGO DETECTION - Disclosed herein are techniques for detecting logos in images or video. In one embodiment, a first logo detection model detects, from an image, candidate regions for determining logos in the image. A feature vector is then extracted from each candidate region and is compared with reference feature vectors stored in a database. The logo corresponding to the best matching reference feature vector is determined to be the logo in the candidate region if the best matching meets a certain criterion. In some embodiments, a second logo detection model trained using synthetic training images is used in combination with the first logo detection model to detect logos in a same image. | 2020-04-30 |
20200134378 | METHOD AND APPARATUS FOR DETECTING OBJECTS OF INTEREST IN AN ENVIRONMENT - A method is provided for generating training data to facilitate automatically locating an object of interest within an image. Methods may include: receiving sensor data including a plurality of images from at least one image sensor; receiving an identification, from a user, of an object visible within an image of the plurality of images, where at least a portion of the object is visible in one or more of the plurality of images; determining a predicted location of the object in the one or more of the remaining images of the plurality of images; identifying the object in the one or more of the remaining images of the plurality of images; and storing the plurality of images including an indication of the object at the object location within the one or more of the plurality of images. | 2020-04-30 |
20200134379 | AUTO-LABELING OF DRIVING LOGS USING ANALYSIS-BY-SYNTHESIS AND UNSUPERVISED DOMAIN ADAPTATION - Acquiring labeled data can be a significant bottleneck in the development of machine learning models that are accurate and efficient enough to enable safety-critical applications, such as automated driving. The process of labeling of driving logs can be automated. Unlabeled real-world driving logs, which include data captured by one or more vehicle sensors, can be automatically labeled to generate one or more labeled real-world driving logs. The automatic labeling can include analysis-by-synthesis on the unlabeled real-world driving logs to generate simulated driving logs, which can include reconstructed driving scenes or portions thereof. The automatic labeling can further include simulation-to-real automatic labeling on the simulated driving logs and the unlabeled real-world driving logs to generate one or more labeled real-world driving logs. The automatically labeled real-world driving logs can be stored in one or more data stores for subsequent training, validation, evaluation, and/or model management. | 2020-04-30 |
20200134380 | Method for Updating Neural Network and Electronic Device - Disclosed are a method for updating a neural network and an electronic device. The method includes: inputting a first image set having tag information into a first depth neural network, and determining a cross entropy loss value of the first image set by using the first depth neural network; inputting a second image set having no tag information separately into the first depth neural network and a second depth neural network, and determining a consistency loss value of the second image set, the first depth neural network and the second depth neural network having the same network structure; updating parameters of the first depth neural network based on the cross entropy loss value and the consistency loss value; and updating parameters of the second depth neural network based on the updated parameters of the first depth neural network. | 2020-04-30 |
20200134381 | AUTOMATED DESIGN TESTING THROUGH DEEP LEARNING - A method is used in evaluating a test subject in computing environments. A first machine learning system generates test subject features. A second machine learning system analyzes the test subject to detect distinguishing features of the test subject. A third machine learning system performs natural language processing on the test subject features to create evaluation information associated with the test subject. A test subject evaluation system provides an evaluation of the test subject based on the distinguishing features and the evaluation information. | 2020-04-30 |
20200134382 | NEURAL NETWORK TRAINING UTILIZING SPECIALIZED LOSS FUNCTIONS - Systems and methods for neural network training utilizing specialized loss functions. An example method comprises: receiving, by a computer system, a training dataset comprising a plurality of images, wherein each image of the training dataset is associated with an identifier of a class of a set of classes; computing, by a neural network, a plurality of feature vectors, wherein each feature vector of the plurality of feature vectors represents an image of the training dataset in a space of image features; computing, for the training dataset, a value of a loss function reflecting a plurality of probabilities, wherein each probability of the plurality of probabilities characterizes a hypothesis associating an image of the training dataset with a class associated with the image by the training dataset, wherein the loss function further reflects a plurality of distances, wherein each distance of the plurality of distances is computed in the space of image features between a feature vector representing an image of the training dataset and a center of a class associated with the image by the training dataset; and adjusting one or more parameters of the neural network based on the value of the loss function. | 2020-04-30 |
20200134383 | GENERATIVE MODEL TRAINING AND IMAGE GENERATION APPARATUS AND METHOD - An apparatus for training an image generative model and an image generating apparatus are provided. The apparatus generates output images from a plurality of input images based on the generative model, extracts depth features from the respective output images based on a depth classification model, calculates a depth loss from the extracted depth features, and trains the generative model based on an overall loss that includes the calculated depth loss. | 2020-04-30 |
20200134384 | ABNORMALITY DETECTION DEVICE - The learning model building unit ( | 2020-04-30 |
20200134385 | DEEP LEARNING MODEL USED FOR IMAGE RECOGNITION AND TRAINING APPARATUS OF THE MODEL AND METHOD THEREOF - Embodiments of this disclosure provide a deep learning model used for image recognition and apparatus and method thereof. The model includes a determination layer configured to determine whether features in feature maps are features of positions where objects of attention are located, and different weights are granted for the positions where the objects of attention are located and other features in performing weight and composition processing on the features. Hence, the model may be guided to be focused on attention features and make correct determination, thereby improving performance and precision of the model. | 2020-04-30 |
20200134386 | APPARATUS AND METHOD FOR CLASSIFYING IMAGE - An image classification apparatus includes an image segmentation module configured to segment a learning image into a plurality of segment images, a primary classification module configured to perform machine learning on a primary classifier using the plurality of segment images, and a secondary classification module configured to calculate a weight value combination for creating a secondary classification estimation value for the learning image from a plurality of primary classification estimation values generated by passing the plurality of segment images to the trained primary classifier, or a machine learning-based learning parameter. | 2020-04-30 |
20200134387 | EVALUATION OF MODELING ALGORITHMS WITH CONTINUOUS OUTPUTS - Certain aspects involve evaluating modeling algorithms whose outputs can impact machine-implemented operating environments. For instance, a computing system generates, from a comparison of a set of estimated attribute values of an attribute to a set of validation attribute values of the attribute, a discretized evaluation dataset with data values in multiple categories. The computing system computes, for a modeling algorithm used to generate the estimated attribute values, an evaluation metric. The computing system provides a host computing system with access to the evaluation metric, one or more modeling outputs generated with the modeling algorithm, or both. Providing one or more of these outputs to the host computing system can facilitate modifying one or more machine-implemented operations. | 2020-04-30 |
20200134388 | Refinement of Machine Learning Engines for Automatically Generating Component-Based User Interfaces - Techniques are disclosed relating to refining, based on user feedback, one or more machine learning engines for automatically generating component-based user interfaces. In various embodiments, a computer system stores template information that defines a plurality of component types and one or more display parameters identified for one or more user interfaces. The computer system may receive a request to generate a user interface, where the request specifies a data set to be displayed. Further, the computer system may automatically generate a user interface, where the generating is performed by one or more machine learning engines that use the template information and the data set as inputs. The computer system may then provide the user interface to one or more users, receive user feedback associated with the user interface, and train at least one of the one or more machine learning engines based on the user feedback. | 2020-04-30 |
20200134389 | ROLLING SHUTTER RECTIFICATION IN IMAGES/VIDEOS USING CONVOLUTIONAL NEURAL NETWORKS WITH APPLICATIONS TO SFM/SLAM WITH ROLLING SHUTTER IMAGES/VIDEOS - A method for correcting rolling shutter (RS) effects is presented. The method includes generating a plurality of images from a camera, synthesizing RS images from global shutter (GS) counterparts to generate training data to train the structure-and-motion-aware convolutional neural network (CNN), and predicting an RS camera motion and an RS depth map from a single RS image by employing a structure-and-motion-aware CNN to remove RS distortions from the single RS image. | 2020-04-30 |
20200134390 | IMPLEMENTING ARTIFICIAL INTELLIGENCE AGENTS TO PERFORM MACHINE LEARNING TASKS USING PREDICTIVE ANALYTICS TO LEVERAGE ENSEMBLE POLICIES FOR MAXIMIZING LONG-TERM RETURNS - A method for implementing artificial intelligence agents to perform machine learning tasks using predictive analytics to leverage ensemble policies for maximizing long-term returns includes obtaining a set of inputs including a set of ensemble policies and a meta-policy parameter, selecting an action for execution within the system environment using a meta-policy function determined based in part on the set of ensemble policies and the meta-policy function parameter, causing the artificial intelligence agent to execute the selected action within the system environment, and updating the meta-policy function parameter based on the execution of the selected action. | 2020-04-30 |
20200134391 | METHOD FOR PREVENTING THE EXTRACTION OF A MACHINE LEARNING MODEL - A method and data processing system for detecting tampering of a machine learning model is provided. The method includes training a machine learning model. During a training operating period, a plurality of input values is provided to the machine learning model. In response to a predetermined invalid input value, the machine learning model is trained that a predetermined output value will be expected. The model is verified that it has not been tampered with by inputting the predetermined invalid input value during an inference operating period. If the expected output value is provided by the machine learning model in response to the predetermined input value, then the machine learning model has not been tampered with. If the expected output value is not provided, then the machine learning model has been tampered with. The method may be implemented using the data processing system. | 2020-04-30 |
20200134392 | DETECTION OF PLANT DISEASES WITH MULTI-STAGE, MULTI-SCALE DEEP LEARNING - In some embodiments, the system is programmed to build from multiple training sets multiple digital models, each for recognizing plant diseases having symptoms of similar sizes. Each digital model can be implemented with a deep learning architecture that classifies an image into one of several classes. For each training set, the system is thus programmed to collect images showing symptoms of one or more plant diseases having similar sizes. These images are then assigned to multiple disease classes. For a first one of the training sets used to build the first digital model, the system is programmed to also include images that correspond to a healthy condition and images of symptoms having other sizes. These images are then assigned to a no-disease class and a catch-all class. Given a new image from a user device, the system is programmed to then first apply the first digital model. For the portions of the new image that are classified into the catch-all class, the system is programmed to then apply another one of the digital models. The system is programmed to finally transmit classification data to the user device indicating how each portion of the new image is classified into a class corresponding to a plant disease or no plant disease. | 2020-04-30 |
20200134393 | NEURAL-NETWORK-BASED CLASSIFICATION DEVICE AND CLASSIFICATION METHOD - Provided is a neural-network-based classification method, including: generating, by a neural network, one or more score vectors corresponding to one or more samples respectively; determining a first subset of the one or more samples according to the one or more score vectors and a first decision threshold, wherein the first subset is associated with a first class; and selecting samples to be re-examined from the one or more samples according to the first subset. | 2020-04-30 |
20200134394 | AGGREGATED STOCHASTIC METHOD FOR PREDICTIVE SYSTEM RESPONSE - An information handling system operating a sensor fusion prediction based automatic adjustment system may comprise sensors measuring influencing attributes comprising information handling system operational values, wherein a subset of the influencing attributes influence one of a plurality of system characteristics, and a memory storing definitions of a user behavior characteristic, a performance mapping characteristic, a power status characteristic, a security profile characteristic, and a policy configuration characteristic. A processor may execute code instructions to apply stochastic prediction to the subset of influencing attribute values to predict a future value of a system characteristic influenced by the subset of influencing attribute values at a future instance in time, determine an adjustment to a policy controlling operational bounds of the system characteristic if the predicted future value of the system characteristic falls outside current policy-defined operating bounds, and automatically perform the policy adjustment before the future instance in time. | 2020-04-30 |
20200134395 | ID ASSOCIATION AND INDOOR LOCALIZATION VIA PASSIVE PHASED-ARRAY AND COMPUTER VISION MOTION CORRELATION - A system and method for combining computer vision information about human subjects within the field-of-view of a computer vision subsystem with RF Angle of Arrival (AoA) information from an RF receiver subsystem to locate, identify, and track individuals and their location. The RF receiver subsystem may receive RF signals emitted by one or more electronic devices (e.g., a mobile phone) carried, held, or otherwise associated with am individual. Further, gestures can be made with the device and they can be detected by the system. | 2020-04-30 |
20200134396 | OBSTACLE DETECTION IN VEHICLE USING A WIDE ANGLE CAMERA AND RADAR SENSOR FUSION - An apparatus includes a primary surround view camera, a supplementary camera, a detection and ranging sensor, and a surround view display. The primary surround view camera is generally placed at a front of a vehicle and provides an operator of the vehicle with a view of the road. The at least one detection and ranging sensor is generally mounted adjacent to the supplementary camera and configured to detect obstacles within a field of view of the supplementary camera. An output of the primary surround view camera is generally used to produce a two-dimensional view of an area around the vehicle and an output of the supplementary camera is (i) reduced to a portion of the field of view of the supplementary camera in which the detection and ranging sensor detected an obstacle and (ii) overlaid on the two-dimensional view of the area around the vehicle to inform the operator of the detected obstacle. | 2020-04-30 |
20200134397 | DEVICE AND METHOD FOR PROCESSING METADATA - A method and an electronic device are disclosed. The method includes obtaining an image, obtaining information of the image, obtaining content information of content included in the image, obtaining related information which relates to the image based on at least one of the information of the image and the content information, and classifying the image into at least one category based on a plurality of defined information/data elements and a relation among the information/data elements and metadata of the image. | 2020-04-30 |
20200134398 | DETERMINING INTENT FROM MULTIMODAL CONTENT EMBEDDED IN A COMMON GEOMETRIC SPACE - Inferring multimodal content intent in a common geometric space in order to improve recognition of influential impacts of content includes mapping the multimodal content in a common geometric space by embedding a multimodal feature vector representing a first modality of the multimodal content and a second modality of the multimodal content and inferring intent of the multimodal content mapped into the common geometric space such that connections between multimodal content result in an improvement in recognition of the influential impact of the multimodal content. | 2020-04-30 |
20200134399 | TRANSFORMING SOURCE DISTRIBUTION TO TARGET DISTRIBUTION USING SOBOLEV DESCENT - Systems, computer-implemented methods, and computer program products for transforming a source distribution to a target distribution. A system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a sampling component that receives a source distribution having a source sample and a target distribution having a target sample. The computer executable components can further comprise an optimizer component that employs a neural network to find a critic that dynamically discriminates between the source sample and the target sample, while constraining a gradient of the neural network. The computer executable components can further comprise a morphing component that generates a first product distribution by morphing the source distribution along the gradient of the neural network to the target distribution. | 2020-04-30 |
20200134400 | FAST COMPUTATION OF A CONVOLUTIONAL NEURAL NETWORK - A computer-implemented method includes obtaining a trained convolutional neural network comprising one or more convolutional layers, each of the one or more convolutional layers comprising a plurality of filters with known filter parameters; pre-computing a reusable factor for each of the one or more convolutional layers based on the known filter parameters of the trained convolutional neural network; receiving input data to the trained convolutional neural network; computing an output of the each of the one or more convolutional layers using a Winograd convolutional operator based on the pre-computed reusable factor and the input data; and determining output data of the trained convolutional network based on the output of the each of the one or more convolutional layers. | 2020-04-30 |
20200134401 | IMAGE PROCESSING OF WEBPAGES - A web detection system processes webpage information and performs automated feature extraction of webpages including machine processable information. In an embodiment, the web detection system determines a subset of webpages having a target characteristic by processing markup language. For a webpage of the subset, the web detection system determines that a first image overlaps at least a portion of a second image in the webpage. The web detection system generates an image of the webpage such that the portion of the second image is obscured by the first image. The web detection system determines a graphical feature of the webpage by processing the image, e.g., using optical character recognition. Responsive to determining that the graphical feature corresponds to graphical features of images of a different set of webpages associated with a target entity, the web detection system determines that the webpage is also associated with the target entity. | 2020-04-30 |
20200134402 | IMAGE PROCESSING APPARATUS, CONTROL METHOD THEREOF AND STORAGE MEDIUM - In processing to thicken a white thin line, the application range thereof is controlled to as to prevent a white thin line not intended by a user from being thickened. The thickening processing is performed for a line that has a density less than or equal to a predetermined density and includes a pixel having attribute information of a drawing object; and not performed for a line that has a density less than or equal to the predetermined density and includes a pixel not having attribute information of the drawing object. | 2020-04-30 |
20200134403 | IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND STORAGE MEDIUM - In conventional color shading (CS) processing for correcting color unevenness with high precision, even an image to preserve pure colors can be corrected to use inks of other colors. Color unevenness is more appropriately corrected to provide a favorable image by properly using pure color preservation information and pure color non-preservation information as color correction information for the CS processing. | 2020-04-30 |
20200134404 | ENCODING PRINTED GRAYSCALE IMAGES - In an example method, a dot pattern of pixels including information to be encoded across an image is mapped to a corresponding subset of the grayscale source pixels corresponding to the image to be printed. A value of a grayscale pixel in the subset of the grayscale source pixels is modified based on based on a predetermined threshold pixel value. The value of the grayscale pixel is decreased in response to detecting that the predetermined threshold pixel value is exceeded. The clipping channel color is used to detect the dot pattern of pixels. The image including the subset of pixels with modified values is printed. | 2020-04-30 |
20200134405 | METHODS AND SYSTEMS FOR HANDLING PRINTING OF LARGE-SIZE OBJECTS - Described herein are methods and devices for printing a large-size object on multiple sheets. The method includes receiving, at a printing device, a print job submitted by a user, wherein the print job includes a large-size object. The large-size object present in the print job is processed by the printing device to ascertain the number and sizes of sheets required for printing the large-size object. Then, the large-size object is printed by the printing device on multiple sheets based on the ascertainment. | 2020-04-30 |
20200134406 | PRINTER SYSTEM, PRINTER, METHOD OF CONTROLLING PRINTER SYSTEM, AND METHOD OF CONTROLLING PRINTER - A printer system includes an information processing device; a printer; and a display apparatus, the information processing device being configured to control the printer and the display apparatus. A topology of the information processing device and the display apparatus includes a first topology in which the display apparatus is directly connected to the information processing device and a second topology in which the display apparatus is connected to the information processing device via the printer. The information processing device includes a processor configured to transmit to the display apparatus instruction data for the display apparatus without adding header information to the instruction data in the first topology; and transmit to the printer instruction data for the display apparatus by adding header information to the instruction data in the second topology. | 2020-04-30 |
20200134407 | IMAGE FORMING APPARATUS AND CONTROL METHOD FOR IMAGE FORMING APPARATUS - An image forming apparatus on which a replaceable container storing a recording material is mounted includes an image forming unit configured to form an image using the recording material, a determination unit configured to determine whether the container satisfies a predetermined condition, an acquisition unit configured to acquire an amount of the recording material used for image formation in a predetermined period and stored in the container determined as a container that satisfies the predetermined condition, a memory configured to accumulate information indicating the amount of the recording material acquired by the acquisition unit, and a prediction unit configured to predict a number of days about replacement of the container, based on the information indicating the amount of the recording material and accumulated in the memory. | 2020-04-30 |
20200134408 | ULTRASONICALLY WELDED LABEL SYSTEMS AND METHODS - Systems and methods of using ultrasonic welding to form labels with RFID tags are disclosed. The methods can be useful for the production of a large volume of labels such as production with roll-to-roll processing. The labels can be useful for consumer products such as garments. The present invention discloses in one embodiment, a label having a first and second printed fabric label layer such that a radio frequency identification (RFID) inlay is disposed between the two printed fabric layers | 2020-04-30 |
20200134409 | TAMPER-PROOF QUALITY MANAGEMENT BARCODE INDICATORS - A tamper-proof barcoded quality indicator operative to provide a machine-readable indication of exceedance of time and temperature thresholds following actuation thereof, including a first barcode including a first colorable area and being machine-readable before exceedance of the time and temperature thresholds, a second barcode including a second colorable area and not being machine-readable before exceedance of the time and temperature thresholds, a coloring agent located at a first location on the indicator, a coloring agent pathway operative to allow the coloring agent to move, at a rate which is at least partially a function of time, from the first location to the first and second colorable areas simultaneously for simultaneous coloring thereof upon exceedance of the time and temperature thresholds, thereby causing the first barcode to become unreadable and at the same time causing the second barcode to become machine-readable, and a tamper-proof actuator element operative to actuate the indicator. | 2020-04-30 |
20200134410 | TAMPER-EVIDENT ITEM AND ITEM VALIDATION SYSTEM AND METHOD - A tamper sensing element is provided which can be applied to a product having a housing or casing with a critical area where the housing or casing can be opened or separated at or along a critical area. The tamper sensing element is operable to determine if the housing or casing has been opened or separated at or along the critical area. The tamper sensing element comprises a sensor for detecting a change in a monitored parameter indicative of the housing or casing having been opened or separated at or along the critical area, a memory for storing product data and tag data; a circuit for updating the memory upon detection of a change in the parameter monitored by the sensor; and means for transmitting information contained in the memory upon being queried by a scanning device. Also disclosed is a system and method for use at security check points for scanning/screening products provided with the tamper sensing element so that a security official/screener can readily determine if a product has potentially been tampered with. | 2020-04-30 |
20200134411 | CONFIGURING A SET OF APPLETS ON A BATTERY-LESS TRANSACTION CARD - A transaction card may power on the transaction card using electric current induced from an interaction of the transaction card with an electromagnetic field. The transaction card may establish a communication with a device. The communication may indicate that the transaction card has powered. The transaction card may receive, from the device, a set of instructions to configure a set of applets on the transaction card after notifying the device that the transaction card has powered on. The set of applets to be configured may be related to completing one or more different transactions. The set of applets to be configured may be different than another set of applets already configured on the transaction card. The transaction card may configure the set of applets on the transaction card according to the set of instructions after receiving the set of instructions. | 2020-04-30 |
20200134412 | Wireless Communications Device with Concealed Value - A wireless communication device is disclosed. The wireless communication device has a near field communications transmitter system. The device has an outer chassis that can be sized, dimensioned and decorated to resemble common household items that lack any apparent value. The wireless communication device can also include an indicia that is invisible to an unaided human eye. | 2020-04-30 |
20200134413 | Method for Fabricating a Smart Card Device - Embodiments of the invention relate to processes for fabricating a smart device, e.g. smart card, and configurations for smart card devices with greater reliability and lifespan, and improved finish. In the smart card device comprising of laminated substrate layers interposing a flexible film having conductor pattern thereon, at least one flip chip for operating the smart card device is embedded in a first substrate such that the first substrate provides an encapsulation to the at least one flip chip, wherein the at least one flip chip is arranged at a position in a first vertical plane; and a contact pad, for providing electrical connection when the smart card device is inserted into a smart card reader, is arranged at a position in a second vertical plane, wherein the first vertical plane is non-overlapping with the second vertical plane. The contact pad is projected through a cavity in a second substrate to form a continuous even plane from an outer surface of the laminated substrate layers to the contact pad. | 2020-04-30 |
20200134414 | DETERMINING RATIONALE OF COGNITIVE SYSTEM OUTPUT - According to one or more embodiments of the present invention, a computer-implemented method includes generating, by a cognitive system, an answer for a user-provided query using an analytics algorithm. The answer is based on a set of data sources. The method further includes determining an influence weightage of each data source from the set of data sources. The method further includes generating and presenting a rationale for the answer based on the influence weightage. | 2020-04-30 |
20200134415 | Autoencoder-Based Generative Adversarial Networks for Text Generation - In accordance to embodiments, an encoder neural network is configured to receive a one-hot representation of a real text and output a latent representation of the real text generated from the one-hot representation of the real text. A decoder neural network is configured to receive the latent representation of the real text, and output a reconstructed softmax representation of the real text from the latent representation of the real text, the reconstructed softmax representation of the real text is a soft-text. A generator neural network is configured to generate artificial text based on random noise data. A discriminator neural network is configured to receive the soft-text and receive a softmax representation of the artificial text, and output a probability indicating whether the softmax representation of the artificial text received by the discriminator neural network is not from the generator neural network. | 2020-04-30 |
20200134416 | SYSTEM-ON-A-CHIP INCORPORATING ARTIFICIAL NEURAL NETWORK AND GENERAL-PURPOSE PROCESSOR CIRCUITRY - A circuit system and a method of analyzing audio or video input data that is capable of detecting, classifying, and post-processing patterns in an input data stream. The circuit system may consist of one or more digital processors, one or more configurable spiking neural network circuits, and digital logic for the selection of two-dimensional input data. The system may use the neural network circuits for detecting and classifying patterns and one or more the digital processors to perform further detailed analyses on the input data and for signaling the result of an analysis to outputs of the system. | 2020-04-30 |
20200134417 | CONFIGURABLE PROCESSOR ELEMENT ARRAYS FOR IMPLEMENTING CONVOLUTIONAL NEURAL NETWORKS - Example apparatus disclosed herein include an array of processor elements, the array including rows each having a first number of processor elements and columns each having a second number of processor elements. Disclosed example apparatus also include configuration registers to store descriptors to configure the array to implement a layer of a convolutional neural network based on a dataflow schedule corresponding to one of multiple tensor processing templates, ones of the processor elements to be configured based on the descriptors to implement the one of the tensor processing templates to operate on input activation data and filter data associated with the layer of the convolutional neural network to produce output activation data associated with the layer of the convolutional neural network. Disclosed example apparatus further include memory to store the input activation data, the filter data and the output activation data associated with the layer of the convolutional neural network. | 2020-04-30 |
20200134418 | SPACE UTILIZATION MEASUREMENT AND MODELING USING ARTIFICIAL INTELLIGENCE - Methods and systems for training a multivariate model predicting utilization of a space. One method includes receiving, over a period of time, signals from each of a plurality of mobile devices located in the space and generating, based on the signals received from each of the plurality of mobile devices, a plurality of location data points for each of the plurality of mobile devices, each of the plurality of location data points for a mobile device including a timestamp and a position within the space of the mobile device. The method also includes accessing metadata of the space, and using, with an electronic processor, the plurality location data points for each of the plurality of mobile devices and the metadata of the space to train machine learning engine. In addition, the method includes predicting a utilization of the space using the machine learning engine. | 2020-04-30 |
20200134419 | RECURRENT NEURON IMPLEMENTATION BASED ON MAGNETO-ELECTRIC SPIN ORBIT LOGIC - Techniques are provided for implementing a recurrent neuron (RN) using magneto-electric spin orbit (MESO) logic. An RN implementing the techniques according to an embodiment includes a first MESO device to apply a threshold function to an input signal provided at a magnetization port of the MESO device, and scale the result by a first weighting factor supplied at an input port of the MESO device to generate an RN output signal. The RN further includes a second MESO device to receive the RN output signal at a magnetization port of the second MESO device and generate a scaled previous RN state value. The scaled previous state value is a scaled and time delayed version of the RN output signal based on a second weighting factor. The RN input signal is a summation of the scaled previous state value of the RN with weighted synaptic input signals provided to the RN. | 2020-04-30 |
20200134420 | MACHINE-BASED PREDICTION OF VISITATION CAUSED BY VIEWING - For machine-based prediction of visitation, a graph of paired mobile devices is formed where the edges for each pair are based on similarities in visitation and/or metadata for the devices. A machine-learned network embeds the visitation and metadata information, which is used to indicate the similarity. Since the trace data used to show access may be sparse, another machine-learned network completes the route based on routes used by similar devices. Another machine-learned network recommends effectiveness of content based on routes, the graph, metadata, and/or other information. The recommendation is based on training using counterfactual and/or other causal modeling. | 2020-04-30 |
20200134421 | ASSURANCE OF POLICY BASED ALERTING - Embodiments provide for assuring policy based alerting, via clustering, via a first neural network, operational data reported from a network into a plurality of anomalies organized into several clusters; correlating, via the first neural network, alerts received from devices in the network according to the several clusters; determining, via the second neural network, anomaly impacts in the several clusters from the filtered alerts; in response to determining that the anomaly impacts for a first cluster exceed an alerting threshold: identifying a first shared node in the first cluster; identifying a second cluster including a second shared node matching the first shared node that has not been determined to exceed the alerting threshold; and transmitting an alert for the first cluster and the second cluster; and in response to receiving a response to the alert, updating, via the second neural network, the first neural network. | 2020-04-30 |
20200134422 | RELATION EXTRACTION FROM TEXT USING MACHINE LEARNING - A first neural network is operated on a processor and a memory to encode a first natural language string into a first sentence encoding including a set of word encodings. Using a word-based attention mechanism with a context vector, a weight value for a word encoding within the first sentence encoding is adjusted to form an adjusted first sentence encoding. Using a sentence-based attention mechanism, a first relationship encoding corresponding to the adjusted first sentence encoding is determined. An absolute difference between the first relationship encoding and a second relationship encoding is computed. Using a multi-layer perceptron, a degree of analogical similarity between the first relationship encoding and a second relationship encoding is determined. | 2020-04-30 |
20200134423 | DATACENTER LEVEL UTILIZATION PREDICTION WITHOUT OPERATING SYSTEM INVOLVEMENT - Embodiments use a hierarchy of machine learning models to predict datacenter behavior at multiple hardware levels of a datacenter without accessing operating system generated hardware utilization information. The accuracy of higher-level models in the hierarchy of models is increased by including, as input to the higher-level models, hardware utilization predictions from lower-level models. The hierarchy of models includes: server utilization models and workload/OS prediction models that produce predictions at a server device-level of a datacenter; and also top-of-rack switch models and backbone switch models that produce predictions at higher levels of the datacenter. These models receive, as input, hardware utilization information from non-OS sources. Based on datacenter-level network utilization predictions from the hierarchy of models, the datacenter automatically configures its hardware to avoid any predicted over-utilization of hardware in the datacenter. Also, the predictions from the hierarchy of models can be used to detect anomalies of datacenter hardware behavior. | 2020-04-30 |
20200134424 | SYSTEMS AND METHODS FOR DOMAIN ADAPTATION IN NEURAL NETWORKS USING DOMAIN CLASSIFIER - A domain adaptation module is used to optimize a first domain derived from a second domain using respective outputs from respective parallel hidden layers of the domains. | 2020-04-30 |
20200134425 | SYSTEMS AND METHODS FOR DOMAIN ADAPTATION IN NEURAL NETWORKS USING CROSS-DOMAIN BATCH NORMALIZATION - A domain adaptation module is used to optimize a first domain derived from a second domain using respective outputs from respective parallel hidden layers of the domains. | 2020-04-30 |
20200134426 | AUTONOMOUS SYSTEM INCLUDING A CONTINUALLY LEARNING WORLD MODEL AND RELATED METHODS - An autonomous or semi-autonomous system includes a temporal prediction network configured to process a first set of samples from an environment of the system during performance of a first task, a controller configured to process the first set of samples from the environment and a hidden state output by the temporal prediction network, a preserved copy of the temporal prediction network, and a preserved copy of the controller. The preserved copy of the temporal prediction network and the preserved copy of the controller are configured to generate simulated rollouts, and the system is configured to interleave the simulated rollouts with a second set of samples from the environment during performance of a second task to preserve knowledge of the temporal prediction network for performing the first task. | 2020-04-30 |
20200134427 | METHOD OF OUTPUTTING PREDICTION RESULT USING NEURAL NETWORK, METHOD OF GENERATING NEURAL NETWORK, AND APPARATUS THEREFOR - A method of generating a second neural network model according to an example embodiment includes: inputting unlabeled input data to a first neural network model; obtaining prediction results corresponding to the unlabeled input data based on the first neural network model; and generating a second neural network model based on the prediction results of the first neural network model and a degree of distribution of the prediction results. | 2020-04-30 |
20200134428 | SELF-ATTENTIVE ATTRIBUTED NETWORK EMBEDDING - Methods and systems for determining a network embedding include training a network embedding model using training data that includes topology information for networks and attribute information relating to vertices of the networks. An embedded representation is generated using the trained network embedding model to represent an input network, with associated attribute information, in a network topology space. A machine learning task is performed using the embedded representation as input to a machine learning model. | 2020-04-30 |
20200134429 | COMPUTER ARCHITECTURE FOR MULTIPLIER-LESS MACHINE LEARNING - A computer architecture for multiplier-less machine learning is disclosed. According to some aspects, a neural network apparatus include processing circuitry and memory. The processing circuitry accesses a plurality of weights for a neural network layer and an input vector for the neural network layer, the input vector comprising a plurality of data values. The processing circuitry provides the plurality of weights and the input vector to an addition layer. The addition layer generates data value-weight pairs and, for each data value-weight pair, creates an input block comprising a sum of the data value and the weight. The processing circuitry sorts the input blocks generated by the addition layer. The processing circuitry cancels any opposite signed input blocks from the sorted input blocks to generate a set of blocks. The processing circuitry outputs a K | 2020-04-30 |
20200134430 | ANALYZING APPARATUS, ANALYSIS METHOD AND ANALYSIS PROGRAM - The analyzing apparatus: generates first internal data; converts a position of first feature data in a feature space, based on the first internal data and a second learning parameter; reallocates, based on a result of first conversion and the first feature data, the first feature data to a position obtained through the conversion in the feature space; calculates a predicted value of a hazard function of analysis time in a case where the first feature data is given, based on a result of reallocation and a third learning parameter; optimizes the first to third learning parameters, based on a response variable and a first predicted value; generates second internal data, based on second feature data and the optimized first learning parameter; converts a position of the second feature data in the feature space, based on the second internal data and the optimized second learning parameter; and calculates importance data. | 2020-04-30 |
20200134431 | CALCULATION UNIT, CALCULATION SYSTEM AND CONTROL METHOD FOR CALCULATION UNIT - A calculation unit, a calculation system and a control method for calculation unit are provided. The calculation unit includes a data interface configured to be connected to a data bus; a data storage component configured to store target data received by the data interface; a first configuration interface configured to receive channel information, where the channel information is used to indicate a quantity of channels included in the data interface; and an operation control component. The operation control component is configured to perform following operations: determining that the data interface includes M channels according to the channel information; obtaining the target data from the data storage component, where the target data includes sub-data corresponding to each channel in the M channels, and performing a data operation according to the sub-data corresponding to each channel in the M channels. | 2020-04-30 |
20200134432 | Low Latency Long Short-Term Memory Inference with Sequence Interleaving - Systems, apparatuses, and methods for implementing a low latency long short-term memory (LSTM) machine learning engine using sequence interleaving techniques are disclosed. A computing system includes at least a host processing unit, a machine learning engine, and a memory. The host processing unit detects a plurality of sequences which will be processed by the machine learning engine. The host processing unit interleaves the sequences into data blocks and stores the data blocks in the memory. When the machine learning engine receives a given data block, the machine learning engine performs, in parallel, a plurality of matrix multiplication operations on the plurality of sequences in the given data block and a plurality of coefficients. Then, the outputs of the matrix multiplication operations are coupled to one or more LSTM layers. | 2020-04-30 |
20200134433 | INTEGRATED CIRCUIT - An integrated circuit includes a data storage circuit, a weight storage circuit, and an operation circuit. The data storage circuit stores, according to a preset control instruction, gray-scale value data of an image in a first preset time period; the weight storage circuit stores a data weight corresponding to the gray-scale value data in the first preset time period; the operation circuit carries out an operation on the gray-scale value data and the data weight in the first preset time period and outputs first data; the data storage circuit further stores the first data according to the preset control instruction in a second preset time period; the weight storage circuit further stores the data weight corresponding to the first data in the second preset time period; the operation circuit carries out an operation on the first data and the data weight in the second time period and outputs second data. | 2020-04-30 |
20200134434 | ARITHMETIC PROCESSING DEVICE, LEARNING PROGRAM, AND LEARNING METHOD - An arithmetic processing device includes an arithmetic circuit; a register storing operation output data; a statistics acquisition circuit generating, from subject data being either the operation output data or normalization subject data, a bit pattern indicating a position of a leftmost set bit for positive number or a position of a leftmost zero bit for negative number of the subject data, the leftmost bit being a bit different from a sign bit; and a statistics aggregation circuit generating either positive or negative statistical information, or both positive and negative statistical information, by separately adding up a first number at respective bit positions of the leftmost set bit indicated by the bit pattern of each of a plurality of subject data having a positive sign bit and a second number of at respective bit positions of the leftmost zero bit indicated by the bit pattern of each of a plurality of subject data having a negative sign bit. | 2020-04-30 |
20200134435 | COMPUTATION APPARATUS, CIRCUIT AND RELEVANT METHOD FOR NEURAL NETWORK - The present disclosure relates to a computation apparatus for a neural network. The computation apparatus includes a first processing unit and a second processing unit. The first processing unit is configured to perform a first computation on k1 number of input feature data according to a size of a computation window to obtain an intermediate result, where a size of the computation window is k1×k2, and k1 and k2 are positive integers. The second processing unit is configured to perform a second computation on k2 number of intermediate results output by the first processing unit according to the size of the computation window to obtain a computation result. | 2020-04-30 |
20200134436 | CONVOLUTIONAL NEURAL NETWORK - A image recognition system includes a first convolution layer, a pooling layer, a second convolution layer, a crossbar circuit having a plurality of input lines, at least one output line intersecting with the input lines, and a plurality of weight elements that are provided at intersection points between the input lines and the output line, weights each input value input to the input lines to output to the output line, and a control portion that selects from convolution operation results of the first convolution layer, an input value needed to acquire each pooling operation result needed to perform second filter convolution operation at each shift position in the second convolution layer, and inputs the input value selected to the input lines. | 2020-04-30 |
20200134437 | IN-MEMORY DATA POOLING FOR MACHINE LEARNING - A method comprises a first block of memory cells to store an input array, and a second block of memory cells. Pooling circuitry is operatively coupled to the first block of memory cells to execute in-place pooling according to a function over the input array to generate an array of output values. Writing circuitry is operatively coupled to the second block to store the array of output values in the second block of memory cells. Analog sensing circuitry is coupled to the first block of memory cells to generate analog values for the input array, wherein the pooling circuitry receives the analog values as inputs to the function. The writing circuitry operatively coupled to the second block is configured to store an analog level in each cell of the second block for the array of output values. | 2020-04-30 |
20200134438 | DYNAMIC RANGE AND LINEARITY OF RESISTIVE ELEMENTS FOR ANALOG COMPUTING - A resistive network include multiple resistive units; each resistive unit is made up of multiple resistive elements, which can be arranged in a parallel configuration. Each of the resistive elements can be programmable (e.g., switched on or off, or set to one of multiple resistance values). Furthermore, a method of analog computing includes configuring multiple resistive elements in each of multiple resistive units and configuring the resistive units into a network. The configuration of the resistive elements can be, for example, arranging them into a parallel combination. The method further includes programming each resistive unit, for example, by switching individual resistive elements into, or out of, the parallel combination. | 2020-04-30 |