NEURONAL NETWORKS

VolverGo back

Resultados 332 results. LastUpdate Updated on 19/06/2021 [02:00:00] pdf PDF xls XLS

Solicitudes publicadas en los últimos 60 días / Applications published in the last 60 days



Page1/14 nextPage   results/page


DEEP LEARNING SEISMIC ATTRIBUTE FAULT PREDICTIONS

Publication No.: US2021181362A1 17/06/2021

Applicant:

LANDMARK GRAPHICS CORP [US]

Absstract of: US2021181362A1

This disclosure presents a fault prediction system using a deep learning neural network, such as a convolutional neural network. The fault prediction system utilizes as input seismic data, and then derives various seismic attributes from the seismic data. In various aspects, the seismic attributes can be normalized and have importance coefficients determined. A sub-set of seismic attributes can be selected to reduce computing resources and processing time. The deep learning neural network can utilize the seismic data and seismic attributes to determine parameterized results representing fault probabilities. The fault prediction system can utilize the fault probabilities to determine fault predictions which can be represented as a predicted new seismic data, such as using a three-dimensional image.

traducir

CLASSIFICATION USING HYPER-OPINIONS

Publication No.: US2021182631A1 17/06/2021

Applicant:

BAKER SUZANNE M [US]
CAMPBELL MATTHEW L [US]
PARSONS THOMAS T [US]

Absstract of: US2021182631A1

Systems, devices, methods, and computer-readable media for determining a hyper-opinion classification of an object. A method can include receiving data of an object to be classified, and determining, using a neural network, a hyper-opinion classification of the object including an indication of the probabilities of base classes and composite classes that are “or” combinations of proper subsets of the base classes.

traducir

DISTRIBUTED TRAINING OF NEURAL NETWORK MODELS

Publication No.: US2021182660A1 17/06/2021

Applicant:

SOUNDHOUND INC [US]

Absstract of: US2021182660A1

Systems and methods for distributed training of a neural network model are described. Various embodiments include a master device and a slave device. The master device has a first version of the neural network model. The slave device is communicatively coupled to a first data source and the master device, and the first data source is inaccessible by the master device, in accordance with one embodiment. The slave device is remote from the master device. The master device is configured to output first configuration data for the neural network model based on the first version of the neural network model. The slave device is configured to use the first configuration data to instantiate a second version of the neural network model. The slave device is configured to train the second version of the neural network model using data from the first data source and to output second configuration data for the neural network model. The master device is configured to use the second configuration data to update parameters for the first version of the neural network model.

traducir

PREDICTING SUBJECT BODY POSES AND SUBJECT MOVEMENT INTENT USING PROBABILISTIC GENERATIVE MODELS

Publication No.: US2021183073A1 17/06/2021

Applicant:

QUALCOMM INC [US]

US_2020160535_A1

Absstract of: US2021183073A1

Certain aspects of the present disclosure are directed to methods and apparatus for predicting subject motion using probabilistic models. One example method generally includes receiving training data comprising a set of subject pose trees. The set of subject pose trees comprises a plurality of subsets of subject pose trees associated with an image in a sequence of images, and each subject pose tree in the subset indicates a location along an axis of the image at which each of a plurality of joints of a subject is located. The received training data may be processed in a convolutional neural network to generate a trained probabilistic model for predicting joint distribution and subject motion based on density estimation. The trained probabilistic model may be deployed to a computer vision system and configured to generate a probability distribution for the location of each joint along the axis.

traducir

DISTANCE METRICS AND CLUSTERING IN RECURRENT NEURAL NETWORKS

Publication No.: US2021182681A1 17/06/2021

Applicant:

INAIT SA [CH]

Absstract of: US2021182681A1

Distance metrics and clustering in recurrent neural networks. For example, a method includes determining whether topological patterns of activity in a collection of topological patterns occur in a recurrent artificial neural network in response to input of first data into the recurrent artificial neural network, and determining a distance between the first data and either second data or a reference based on the topological patterns of activity that are determined to occur in response to the input of the first data.

traducir

REALISTIC NEURAL NETWORK BASED IMAGE STYLE TRANSFER

Publication No.: US2021183033A1 17/06/2021

Applicant:

SNAP INC [US]

US_10891723_B1

Absstract of: US2021183033A1

A mobile device can implement a neural network-based style transfer scheme to modify an image in a first style to a second style. The style transfer scheme can be configured to detect an object in the image, apply an effect to the image, and blend the image using color space adjustments and blending schemes to generate a realistic result image. The style transfer scheme can further be configured to efficiently execute on the constrained device by removing operational layers based on resources available on the mobile device.

traducir

REINFORCEMENT LEARNING NEURAL NETWORKS GROUNDED IN LEARNED VISUAL ENTITIES

Publication No.: EP3834138A1 16/06/2021

Applicant:

DEEPMIND TECH LTD [GB]

CN_112771542_A

Absstract of: WO2020064994A1

A reinforcement learning neural network system in which internal representations and policies are grounded in visual entities derived from image pixels comprises a visual entity identifying neural network subsystem configured to process image data to determine a set of spatial maps representing respective discrete visual entities. A reinforcement learning neural network subsystem processes data from the set of spatial maps and environmental reward data to provide action data for selecting actions to perform a task.

traducir

METHOD AND SYSTEM FOR PERFORMING OBJECT DETECTION USING A CONVOLUTIONAL NEURAL NETWORK

Publication No.: EP3834134A1 16/06/2021

Applicant:

AVIGILON CORP [CA]

CA_3112265_PA

Absstract of: US2020092463A1

Methods, systems, and techniques for performing object detection using a convolutional neural network (CNN) involve obtaining an image and then processing the image using the CNN to generate a first feature pyramid and a second feature pyramid from the first pyramid. The second pyramid includes an enhanced feature map, which is generated by combining an upsampled feature map and a feature map of the first feature pyramid that has a corresponding or lower resolution of a resolution of the enhanced feature map. The upsampled feature map is generated by upsampling a feature map of the second feature pyramid that is at a shallower position in the CNN than the enhanced feature map. The enhanced feature map is split into channel feature maps of different resolutions, with each of the channel feature maps corresponding to channels of the enhanced feature map. Object detection is performed on the channel feature maps.

traducir

QUANTIZATION METHOD AND APPARATUS FOR NEURAL NETWORK MODEL IN DEVICE

Publication No.: EP3836032A1 16/06/2021

Applicant:

HUAWEI TECH CO LTD [CN]

WO_2020056718_PA

Absstract of: EP3836032A1

A method and an apparatus for quantizing a neural network model in a device is provided. User calibration data is obtained and input into the neural network model to calculate a quantization parameter of each of a plurality of layers of the neural network model, to-be-quantized data is input into the neural network model to quantize input data of each layer by using the quantization parameter of each layer. Because the user calibration data is generated based on data generated by the device used by a user, the quantization parameter can be obtained online, and the quantization parameter matches user data in the device, thereby improving quantization accuracy.

traducir

TWO-PASS END TO END SPEECH RECOGNITION

Publication No.: WO2021113443A1 10/06/2021

Applicant:

GOOGLE LLC [US]

Absstract of: WO2021113443A1

Two-pass automatic speech recognition (ASR) models can be used to perform streaming on-device ASR to generate a text representation of an utterance captured in audio data. Various implementations include a first-pass portion of the ASR model used to generate streaming candidate recognition(s) of an utterance captured in audio data. For example, the first-pass portion can include a recurrent neural network transformer (RNN-T) decoder. Various implementations include a second-pass portion of the ASR model used to revise the streaming candidate recognition(s) of the utterance and generate a text representation of the utterance. For example, the second-pass portion can include a listen attend spell (LAS) decoder. Various implementations include a shared encoder shared between the RNN-T decoder and the LAS decoder.

traducir

SYSTEMS AND METHODS FOR AUTOMATED TRADING

Publication No.: US2021174449A1 10/06/2021

Applicant:

UST GLOBAL INC [US]

Absstract of: US2021174449A1

A system configured to: (a) retrieve structured and unstructured data from one or more external data sources, the structured data including time-series data on a financial instrument and the unstructured data including words; (b) analyze the unstructured data to determine a sentiment measure for the financial instrument; (c) analyze the structured data to obtain a training dataset; (d) train a neural network model with the training dataset such that the neural network can provide a predicted price of the financial instrument for a future timestamp; and (e) provide a decision for managing the financial instrument based at least in part on the sentiment measure for the financial instrument, the predicted price of the financial instrument, and a current holding of the financial instrument.

traducir

DEVICE, METHOD, AND PROGRAM FOR ENHANCING OUTPUT CONTENT THROUGH ITERATIVE GENERATION

Publication No.: US2021174801A1 10/06/2021

Applicant:

SAMSUNG ELECTRONICS CO LTD [KR]

Absstract of: US2021174801A1

A method of improving output content through iterative generation is provided. The method includes receiving a natural language input, obtaining user intention information based on the natural language input by using a natural language understanding (NLU) model, setting a target area in base content based on a first user input, determining input content based on the user intention information or a second user input, generating output content related to the base content based on the input content, the target area, and the user intention information by using a neural network (NN) model, generating a caption for the output content by using an image captioning model, calculating similarity between text of the natural language input and the generated output content, and iterating generation of the output content based on the similarity.

traducir

INTERPRETABLE DEEP MACHINE LEARNING FOR CLINICAL RADIO;OGY

Publication No.: US2021174154A1 10/06/2021

Applicant:

UNIV YALE [US]

WO_2020033594_PA

Absstract of: US2021174154A1

One aspect of the invention provides a computer-implemented method of identifying one or more clinical factors associated with an artificial intelligence prediction. The computer implemented method includes: applying a previously trained deep neural network to one or more images for a subject to produce a prediction, the previously trained deep neural network comprising a plurality of nodes; comparing outputs of the nodes to previously identified patterns of node outputs for a plurality of individual clinical factors; and identifying which of the plurality of individual clinical factors are most correlated with the outputs of the nodes.

traducir

FEATURE FUSION AND DENSE CONNECTION-BASED METHOD FOR INFRARED PLANE OBJECT DETECTION

Publication No.: US2021174149A1 10/06/2021

Applicant:

UNIV XIDIAN [CN]

WO_2020102988_A1

Absstract of: US2021174149A1

A feature fusion and dense connection-based method for infrared plane object detection includes: constructing an infrared image dataset containing an object to be recognized, calibrating a position and class of the object to be recognized in the infrared image dataset, and obtaining an original known label image; dividing the infrared image dataset into a training set and a validation set; performing image enhancement preprocessing on images in the training set, performing feature extraction and feature fusion, and obtaining classification results and bounding boxes through a regression network; calculating a loss function according to the classification results and the bounding boxes in combination with the original known label image, and updating parameter values of a convolutional neural network; repeating the steps to iteratively update the parameters of the convolutional neural network; and processing images in the validation set through the parameters to obtain a final object detection result map.

traducir

METHOD AND ELECTRONIC DEVICE FOR SELECTING DEEP NEURAL NETWORK HYPERPARAMETERS

Publication No.: US2021174210A1 10/06/2021

Applicant:

IND TECH RES INST [TW]

Absstract of: US2021174210A1

A method and an electronic device for selecting deep neural network hyperparameters are provided. In an embodiment of the method, a plurality of testing hyperparameter configurations are sampled from a plurality of hyperparameter ranges of a plurality of hyperparameters. A target neural network model is trained by using a training dataset and the plurality of testing hyperparameter configurations, and a plurality of accuracies corresponding to the plurality of testing hyperparameter configurations are obtained after training for preset epochs. A hyperparameter recommendation operation is performed to predict a plurality of final accuracies of the plurality of testing hyperparameter configurations. A recommended hyperparameter configuration corresponding to the final accuracy having a highest predicted value is selected as a hyperparameter setting for continuing training the target neural network model.

traducir

Computer Vision System for Industrial Equipment Gauge Digitization and Alarms

Publication No.: US2021174131A1 10/06/2021

Applicant:

SCHWARTZ EDWARD L [US]
RICOH CO LTD [JP]

JP_2021093161_A

Absstract of: US2021174131A1

A system for analog gauge monitoring uses a machine learning model for computer vision that is trained using synthetic training data generated based on one or a few images of the gauge being monitored and a geometric model describing the scale and the indicator of the gauge. In some embodiments, the synthetic training data is generated using an image model implemented as a generative adversarial network (GAN) type neural network and trained to modify an image of a given gauge such that the gauge face is preserved while the gauge indicator is added to or removed from the image of the given gauge for any given gauge.

traducir

SPECIALIZED COMPUTER-AIDED DIAGNOSIS AND DISEASE CHARACTERIZATION WITH A MULTI-FOCAL ENSEMBLE OF CONVOLUTIONAL NEURAL NETWORKS

Publication No.: US2021174504A1 10/06/2021

Applicant:

UNIV CASE WESTERN RESERVE [US]
LOUIS STOKES CLEVELAND VETERANS ADMINISTRATION MEDICAL CENTER [US]

Absstract of: US2021174504A1

Embodiments discussed herein facilitate determination of whether lesions are benign or malignant. One example embodiment is a method, comprising: accessing medical imaging scan(s) that are each associated with distinct angle(s) and each comprise a segmented region of interest (ROI) of that medical imaging scan comprising a lesion associated with a first region and a second region; providing the first region(s) of the medical imaging scan(s) to trained first deep learning (DL) model(s) of an ensemble and the second region(s) of the medical imaging scan(s) to trained second DL model(s) of the ensemble; and receiving, from the ensemble of DL models, an indication of whether the lesion is a benign architectural distortion (AD) or a malignant AD.

traducir

ATTENTION-BASED LAYERED NEURAL NETWORK ARCHITECTURE FOR EXPLAINABLE AND HIGH-PERFORMANCE AI PROCESSING

Publication No.: US2021173905A1 10/06/2021

Applicant:

BANK OF AMERICA [US]

Absstract of: US2021173905A1

A system for attention-based layered neural network classification is provided. The system comprises: a sequence of layered neural networks; and a controller configured for controlling data routed through the sequence of layered neural networks, the controller configured to: receive interaction data comprising data features, wherein the data features are distinct characteristics of the interaction data; input data features into the sequence of layered neural networks, wherein each sequential layer of the sequence of layered neural networks comprises a heightened rigor level for at least one of the data features; calculate a relevance score output for at least one of the data features at each layer of the sequence of layered neural networks; and integrate the relevance score output from each layer of the sequence of layered neural networks to generate a total relevance score output.

traducir

SYSTEMS AND METHODS DRIVEN BY LINK-SPECIFIC NUMERIC INFORMATION FOR PREDICTING ASSOCIATIONS BASED ON PREDICATE TYPES

Publication No.: US2021174217A1 10/06/2021

Applicant:

ACCENTURE GLOBAL SOLUTIONS LTD [IE]

CN_112925857_A

Absstract of: US2021174217A1

The present disclosure describes methods and systems to predict predicate metadata parameters in knowledge graphs via neural networks. The method includes receiving a knowledge graph based on a knowledge base including a graph-based dataset. The knowledge graph includes a predicate between two nodes and a set of predicate metadata. The method also includes determining a positive structural score, adjusting each positive structural score based on each corresponding significance parameter, generating a synthetic negative graph-based dataset, determining a negative structural score for each synthetic negative triple of the synthetic negative graph-based dataset, adjusting each negative structural score based on each corresponding significance parameter, determining a significance loss value based on the adjusted positive structural scores and the adjusted negative structural scores, and determining a likelihood score of a link between a third node and a fourth node in the knowledge graph based on the significance loss value.

traducir

NEURAL NETWORK TRAINING USING A DATA FLOW GRAPH AND DYNAMIC MEMORY MANAGEMENT

Publication No.: US2021174190A1 10/06/2021

Applicant:

IBM [US]

Absstract of: US2021174190A1

Processing a neural network data flow graph having a set of nodes and a set of edges. An insertion point is determined for a memory reduction or memory restoration operation. The determination is based on computing tensor timing slacks (TTS) for a set of input tensors; compiling a candidate list (SI) of input tensors, from the set of input tensors, using input tensors having corresponding TTS values larger than a threshold value (thTTS); filtering the SI to retain input tensors whose size meets a threshold value (thS); and determining an insertion point for the operation using the SI based on the filtering. A new data flow graph is generated or an existing one is modified using this process.

traducir

COMPUTATION WITH OPTICAL METAMATERIALS

Publication No.: US2021174186A1 10/06/2021

Applicant:

NEUROPHOS LLC [US]

WO_2021026152_A1

Absstract of: US2021174186A1

Opto-electronic devices can evaluate convolutional neural networks with reduced power consumption and higher speeds using optical metamaterial structures. Methods and systems for convolution of an input vector f with a kernel k can include a first optical element that is adjustable according to the input vector f and a second optical element that is adjustable according to the kernel k, where either or both elements can include adjustable optical metasurfaces. In some approaches, the second optical element is adjustable according to a Fourier transform of the kernel k and is interposed between first and second lenses or volumetric metamaterials implementing Fourier and inverse Fourier transforms, respectively.

traducir

SLOT FILLING WITH CONTEXTUAL INFORMATION

Publication No.: US2021174193A1 10/06/2021

Applicant:

ADOBE INC [US]

CN_112925516_A

Absstract of: US2021174193A1

A system, method and non-transitory computer readable medium for editing images with verbal commands are described. Embodiments of the system, method and non-transitory computer readable medium may include an artificial neural network (ANN) comprising a word embedding component configured to convert text input into a set of word vectors, a feature encoder configured to create a combined feature vector for the text input based on the word vectors, a scoring layer configured to compute labeling scores based on the combined feature vectors, wherein the feature encoder, the scoring layer, or both are trained using multi-task learning with a loss function including a first loss value and an additional loss value based on mutual information, context-based prediction, or sentence-based prediction, and a command component configured to identify a set of image editing word labels based on the labeling scores.

traducir

ELECTRONIC DEVICE AND CONTROLLING METHOD OF ELECTRONIC DEVICE

Publication No.: US2021174137A1 10/06/2021

Applicant:

SAMSUNG ELECTRONICS CO LTD [KR]

Absstract of: US2021174137A1

An electronic device and a controlling method of an electronic device are provided. The electronic device includes identifying whether each of one or more neural network models included in a first external device is suitable for hardware of the electronic device and whether each of the one or more neural network models identified as suitable for the hardware of the electronic device is suitable to replace the neural network models included in the electronic device, based on first device information on a hardware specifications of the electronic device, second device information on a hardware specification of the first external device, first model information on the one or more neural network models included in the first external device, and second model information on the one or more neural network models included in the electronic device.

traducir

VENTRAL-DORSAL NEURAL NETWORKS: OBJECT DETECTION VIA SELECTIVE ATTENTION

Publication No.: US2021174083A1 10/06/2021

Applicant:

ANCESTRY COM OPERATIONS INC [US]

CN_112805717_A

Absstract of: US2021174083A1

Embodiments described herein relate generally to a methodology of efficient object classification within a visual medium. The methodology utilizes a first neural network to perform an attention based object localization within a visual medium to generate a visual mask. The visual mask is applied to the visual medium to generate a masked visual medium. The masked visual medium may be then fed into a second neural network to detect and classify objects within the visual medium.

traducir

HARDWARE-ACCELERATED OPERATION OF ARTIFICIAL NEURAL NETWORKS

Nº publicación: US2021174108A1 10/06/2021

Applicant:

BOSCH GMBH ROBERT [DE]

CN_112926734_A

Absstract of: US2021174108A1

A method for operating an artificial neural network (ANN) on a hardware platform. The ANN is designed to ascertain confidences with which input data are to be assigned to N discrete classes. The hardware platform includes a dedicated unit which forms from a list of M>N confidences expanded confidences by encoding into each confidence an identification number of its place in the list, and numerically sorts the expanded confidences. The unit is fed confidences 1, . . . , M−N, which have the minimal representable value, and confidences M−N+1, . . . , M, which correspond to the N discrete classes, and/or it is ensured that those confidences fed to the unit that correspond to one of the N discrete classes have a value higher than the minimal representable value. A ranking of the classes ordered according to confidences, to which the input data are to be assigned, is ascertained from the first N of the numerically sorted expanded confidences.

traducir

Page1/14 nextPage results/page

punteroimgGo back