REDES NEURONALES

VolverVolver

Resultados 93 resultados LastUpdate Última actualización 29/09/2022 [17:45:00] pdf PDF xls XLS

Solicitudes publicadas en los últimos 30 días / Applications published in the last 30 days



Página1 de 4 nextPage   por página


Graph Neural Networks for datasets with heterophily

NºPublicación: GB2605218A 28/09/2022

Solicitante:

ADOBE INC [US]

Resumen de: GB2605218A

The application relates to training classifier graph neural networks to better handle heterophilic datasets, such as user profiles. In one disclosure, processing a dataset with a graph neural network begins by defining prior belief vectors for nodes in a graph data structure. Then, a compatibility-guided propagation from the prior belief vectors is performed using a compatibility matrix that models a probability of nodes of different classes being connected. The graph neural network predicts a class label for a node based on the compatibility-guided propagation and a characteristic of a node within the neighbourhood of the node of interest. In an alternative disclosure, a graph neural network includes a graph data structure which models a dataset. A neural network is used to generate a belief vector indicating a probability of a class label for a node of the graph data structure. A compatibility matrix modifies the belief vector based on belief vectors of nodes in the node’s neighbourhood. A loss value is calculated based on a co-training loss and a regulation value that keeps the compatibility matrix’s rows centred around zero. Compatibility matrix parameters are updated using the loss value. In another disclosure, a dataset is modelled using a graph neural network having a compatibility matrix that models a probability of nodes of different classes being connected.

traducir

METHODS AND ELECTRONIC DEVICES FOR DETECTING OBJECTS IN SURROUNDINGS OF A SELF-DRIVING CAR

NºPublicación: EP4064127A1 28/09/2022

Solicitante:

YANDEX SELF DRIVING GROUP LLC [RU]

RU_2767831_C1

Resumen de: EP4064127A1

A method (1000) and electronic device (210) for detecting an object are disclosed. The method (1000) includes generating a cluster of points representative of the surroundings of the SDC (220), generating by a first Neural Network (NN) a first feature vector based on the cluster indicative of a local context of the given object in the surroundings of the SDC (220), generating by a second NN second feature vectors for respective points from the cluster based on a portion of the point cloud, where a given second feature vector is indicative of the local and global context of the given object, generating by the first NN a third feature vector for the given object based on the second feature vectors representative of the given object, and generating by a third NN a bounding box around the given object using the third feature vector.

traducir

METHOD AND APPARATUS FOR NEURAL NETWORK MODEL COMPRESSION/DECOMPRESSION

NºPublicación: EP4062320A1 28/09/2022

Solicitante:

TENCENT AMERICA LLC [US]

JP_2022525897_A

Resumen de: US2021159912A1

Aspects of the disclosure provide methods and apparatuses for neural network model compression/decompression. In some examples, an apparatus for neural network model decompression includes receiving circuitry and processing circuitry. The processing circuitry decodes, from a bitstream corresponding to a representation of a neural network, at least a syntax element to be applied to multiple blocks in the neural network. Then, the processing circuitry reconstructs, from the bitstream, weight coefficients in the blocks based on the syntax element.

traducir

ARCHITECTURE FOR AN EXPLAINABLE NEURAL NETWORK

NºPublicación: EP4062330A1 28/09/2022

Solicitante:

UMNAI LTD [MT]

WO_2021099338_A1

Resumen de: WO2021099338A1

An architecture for an explainable neural network may implement a number of layers to produce an output. The input layer may be processed by both a conditional network and a prediction network. The conditional network may include a conditional layer, an aggregation layer, and a switch output layer. The prediction network may include a feature generation and transformation layer, a fit layer, and a value output layer. The results of the switch output layer and value output layer may be combined to produce the final output layer. A number of different possible activation functions may be applied to the final output layer depending on the application. The explainable neural network may be implementable using both general purpose computing hardware and also application specific circuitry including optimized hardware only implementations. Various embodiments of XNNs are described that extend the functionality to different application areas and industries.

traducir

METHOD AND APPARATUS WITH DECODING IN NEURAL NETWORK FOR SPEECH RECOGNITION

NºPublicación: US2022301578A1 22/09/2022

Solicitante:

SAMSUNG ELECTRONICS CO LTD [KR]

CN_115116436_PA

Resumen de: US2022301578A1

A decoding method, the method including: receiving an input sequence corresponding to an input speech at a current time; and in a neural network (NN) for speech recognition, generating an encoded vector sequence by encoding the input sequence, determining reuse tokens from candidate beams of two or more previous times by comparing the candidate beams of the previous times, and decoding one or more tokens subsequent to the reuse tokens based on the reuse tokens and the encoded vector sequence.

traducir

SECURITY OPERATIONS OF PARKED VEHICLES

NºPublicación: US2022301318A1 22/09/2022

Solicitante:

MICRON TECHNOLOGY INC [US]

CN_112406787_A

Resumen de: US2022301318A1

Systems, methods and apparatus of vehicle security operations during parking. For example, a vehicle includes: a proximity sensor configured to detect presence of an object approaching the vehicle when the vehicle is in a parking state; at least one camera configured to monitor surroundings of the vehicle when the vehicle is in the parking state; and an artificial neural network configured to extract identification information of the object from images generated by the camera and determine a security classification of the presence of the object. The identification information is stored in the vehicle and/or transmitted to a server or a mobile device, in response to the security classification being in a predefined category.

traducir

IMAGE COLORIZATION USING MACHINE LEARNING

NºPublicación: US2022301227A1 22/09/2022

Solicitante:

GOOGLE LLC [US]

KR_20220009456_PA

Resumen de: US2022301227A1

Implementations described herein relate to methods, systems, and computer-readable media to train and use a machine-learning model to colorize a grayscale image that depicts a person. In some implementations, a computer-implemented method includes receiving the grayscale image. The method further includes generating a colorized image based on the grayscale image as output of a trained convolutional neural network (CNN) by providing the grayscale image as input to the trained CNN. In some implementations, the trained CNN performs part segmentation to detect one or more parts of the person and colorizes the grayscale image.

traducir

METHODS, SYSTEMS AND COMPUTER MEDIUM FOR SCENE-ADAPTIVE FUTURE DEPTH PREDICTION IN MONOCULAR VIDEOS

NºPublicación: US2022301211A1 22/09/2022

Solicitante:

LIU HUAN [CA]
CHI ZHIXIANG [CA]
YU YUANHAO [CA]
WANG YANG [CA]
TANG JIN [CA]

WO_2022193866_PA

Resumen de: US2022301211A1

Systems, methods and computer-readable medium for predicting a depth for a video frame are disclosed. An example method may include steps of: receiving a plurality of training data, each comprising a set of consecutive video frames and a depth representation of a subsequent video frame to the consecutive video frames; receiving a pre-trained neural network model fθ having a plurality of weights θ; while the pre-trained neural network model fθ has not converged: computing a plurality of second weights θi′, based on each set of consecutive video frames, and updating the plurality of weights θ, based on the plurality of training data and the plurality of second weights θi′; receiving a plurality of new consecutive video frames with consecutive timestamps; and predicting a depth representation of video frame immediately subsequent to the new consecutive video frames based on the updated plurality of weights θ.

traducir

UNCERTAINTY MAPS FOR DEEP LEARNING ELETRICAL PROPERTIES TOMOGRAPHY

NºPublicación: US2022301687A1 22/09/2022

Solicitante:

KONINKLIJKE PHILIPS NV [NL]

CN_114097041_PA

Resumen de: US2022301687A1

The present disclosure relates to a method for determining electrical properties, EP, of a target volume (708) in an imaged subject (718). The method comprises: a) training (201) a deep neural network, DNN, using a training dataset, the training dataset comprising training B1 field maps and corresponding EP maps, the training comprising using a monte carlo, MC, dropout of the DNN during the training, resulting in a trained DNN configured for generating EP maps from B1 field maps; b) receiving (203) an input B1 field map of the target volume, and repeatedly generate by the trained DNN from the input B1 field map an EP map, resulting in a set of EP maps, wherein the generating comprises using in each repetition the MC dropout during inference of the DNN; c) combining (205) the set of EP maps for determining an EP map and associated uncertainty map of the input B1 field map.

traducir

Processing of Chroma-Subsampled Video Using Convolutional Neural Networks

NºPublicación: US2022303557A1 22/09/2022

Solicitante:

AVID TECH INC [US]

Resumen de: US2022303557A1

Efficient processing of chroma-subsampled video is performed using convolutional neural networks (CNNs) in which the luma and chroma channels are processed separately. The luma channel is independently convolved and downsampled and, in parallel, the chroma channels are convolved and then merged with the downsampled luma to generate encoded chroma-subsampled video. Further processing of the encoded video that involves deconvolution and upsampling, splitting into two sets of channels, and further deconvolutions and upsampling is used in CNNs to generate decoded chroma-subsampled video in compression-decompression applications, to remove noise from chroma-subsampled video, or to upsample chroma-subsampled video to RGB 444 video. CNNs with separate luma and chroma processing in which the further processing includes additional convolutions and downsampling may be used for object recognition and semantic search in chroma-subsampled video.

traducir

DIRECT POLICY OPTIMIZATION FOR MEETING ROOM COMFORT CONTROL AND ENERGY MANAGEMENT

NºPublicación: US2022299233A1 22/09/2022

Solicitante:

JOHNSON CONTROLS TECH CO [US]

Resumen de: US2022299233A1

A method for controlling temperature in a building zone to increase comfort and energy efficiency is shown. The method includes receiving historical data, the historical data indicative of the temperature and occupancy of the building zone during one or more historical states. The method includes training a system model to represent a dynamic response of the building zone based on the historical data. The method includes determining a control law by optimizing a policy function implemented as a neural network configured to process the trained system model. The method includes performing online control of the building zone using the control law.

traducir

MODULAR MACHINE LEARNING MODELS FOR DENOISING IMAGES AND SYSTEMS AND METHODS FOR USING SAME

NºPublicación: US2022301112A1 22/09/2022

Solicitante:

MICRON TECHNOLOGY INC [US]

Resumen de: US2022301112A1

In some examples, a machine learning model may be trained to denoise an image. In some examples, the machine learning model may identify noise in an image of a sequence based at least in part, on at least one other image of the sequence. In some examples, the machine learning model may include a recurrent neural network. In some examples, the machine learning model may have a modular architecture including one or more building units. In some examples, the machine learning model may have a multi-branch architecture. In some examples, the noise may be identified and removed from the image by an iterative process.

traducir

METHOD OF GENERATING MULTI-LAYER REPRESENTATION OF SCENE AND COMPUTING DEVICE IMPLEMENTING THE SAME

NºPublicación: WO2022197084A1 22/09/2022

Solicitante:

SAMSUNG ELECTRONICS CO LTD [KR]

Resumen de: WO2022197084A1

The present disclosure relates to the field of artificial intelligence (AI) and neural rendering, and particularly to a method of generating a multi-layer representation of a scene using neural networks trained in an end-to-end fashion and to a computing device implementing the method. The method of generating a multi-layer representation of a scene includes: obtaining a pair of images of the scene, the pair of the images comprising a reference image and a source image; performing a reprojection operation on the pair of images to generate a plane-sweep volume; predicting, using a geometry network, a layered structure of the scene based on the plane-sweep volume; and estimating, using a coloring network, color values and opacity values for the predicted layered structure of the scene to obtain the multi-layer representation of the scene; wherein the geometry network and the coloring network are trained in end-to-end manner.

traducir

AUTOMATED EMPATHETIC ASSESSMENT OF CANDIDATE FOR JOB

NºPublicación: US2022300736A1 22/09/2022

Solicitante:

FUJITSU LTD [JP]

EP_4060581_PA

Resumen de: US2022300736A1

In an embodiment, operations include extracting first information about a first set of features of a first candidate, from a document or profile information of the first candidate. Second information about a second set of features, corresponding to the first set of features, is extracted from one or more databases. The second set of features is associated with a population of candidates with at least one demographic parameter same as that of the first candidate. A third set of features is determined based on difference of corresponding features from the first set of features and the second set of features. A pre-trained neural network model is applied on the third set of features to determine a set of weights associated with the third set of features. An empathy score of the first candidate is determined based on the set of weights. The empathy score of the first candidate is rendered.

traducir

NEURAL NETWORK OF PREDICTING IMAGE DEFINITION, TRAINING METHOD AND PREDICTION METHOD

NºPublicación: US2022300767A1 22/09/2022

Solicitante:

BOE TECHNOLOGY GROUP CO LTD [CN]

CN_112926496_A

Resumen de: US2022300767A1

The present application disclose a neural network of predicting image definition, a training method and a prediction method. The training method includes: obtaining an image set and definition labels of some images in the image set, thereby obtaining image samples with the definition labels and to-be-expanded images except for the image samples; and extracting definition features of at least some images in the image set, obtaining definition labels of at least some images in the to-be-expanded images according to the extracted definition features, correcting the definition labels of the at least some images in the to-be-expanded images to expand the image samples, and using the image samples to train the neural network of predicting image definition, thereby obtaining a trained neural network.

traducir

CLASSIFICATION SYSTEM AND METHOD OF INFORMATION IN IMAGE

NºPublicación: US2022300771A1 22/09/2022

Solicitante:

INVENTEC PUDONG TECH CORP [CN]
INVENTEC CORP [TW]

Resumen de: US2022300771A1

A classification method of information in image comprises receiving an input image and generating a plurality of shared feature maps by a convolutional neural network; generating a plurality of attention maps according to the plurality of shared feature maps by an attention network; selecting at least two of the plurality of attention maps to perform a fusion operation to generate a fusion map by a fusion circuit; and generating a classification result according to the fusion map by a classifier.

traducir

CONTEXT-AWARE ENTITY LINKING FOR KNOWLEDGE GRAPHS

NºPublicación: US2022300831A1 22/09/2022

Solicitante:

NEC LABORATORIES EUROPE GMBH [DE]

Resumen de: US2022300831A1

A machine learning model includes a context transformer and a decision head. The context transformer is a neural network of self-attention layers. The model makes a link prediction for a query embedding. Input embeddings are received at inputs of the context transformer. The input embeddings have: a query embedding set, the query embedding set comprising a subject embedding, object embedding, and relation embedding, one of the subject embedding, the object embedding, and the relation embedding being the query embedding; and knowledge graph embeddings. A first self-attention layer generates an attention score for each of the input embeddings. A final layer of the context transformer generates the link prediction for the query embedding and an output associated with each of the input embeddings. The decision head combines the attention score and the output for each of the input embeddings to determine a significance score for each of the input embeddings.

traducir

METHOD AND APPARATUS FOR DETERMINING A WAVEFORM MODULATION VECTOR INDICATIVE OF A MODULATED CARRIER SIGNAL USING AN ARTIFICIAL NEURAL NETWORK

NºPublicación: WO2022194516A1 22/09/2022

Solicitante:

FRAUNHOFER GES FORSCHUNG [DE]

EP_4060367_PA

Resumen de: WO2022194516A1

The present disclosure relates to an estimation of a waveform modulation vector (440) indicative of a modulated carrier signal with signal characteristics. A code vector (420) representing real values encoding the signal characteristics of the waveform modulation vector is provided to an artificial neural network (430). The code vector has a dimension smaller than a dimension of the waveform modulation vector (440). The artificial neural network (430) is configured to output the waveform modulation vector (440) based on the code vector (420).

traducir

SYSTEMS AND METHODS FOR TRAINING AND EXECUTING A NEURAL NETWORK FOR COLLABORATIVE MONITORING OF RESOURCE USAGE

NºPublicación: US2022300813A1 22/09/2022

Solicitante:

CAPITAL ONE SERVICES LLC [US]

US_2020364560_A1

Resumen de: US2022300813A1

Disclosed are systems and methods for training and executing a neural network for collaborative monitoring of resource usage metrics. For example, a method may include receiving user data sets, grouping the user data sets into one or more clusters of user data sets, grouping each of the one or more clusters into a plurality of subclusters, for each of the plurality of subclusters, training the neural network to associate the subcluster with one or more sequential patterns found within the subcluster, grouping the plurality of user data sets into a plurality of teams, receiving a first series of transactions of a first user, inputting the first series of transactions into the trained neural network, classifying, using the trained neural network, the first user into a subcluster among the plurality of subclusters, generating a metric associated with the first series of transactions, generating a recommendation to the first user.

traducir

TECHNIQUES FOR ADAPTIVE GENERATION AND VISUALIZATION OF QUANTIZED NEURAL NETWORKS

NºPublicación: US2022300800A1 22/09/2022

Solicitante:

VIANAI SYSTEMS INC [US]

WO_2022197615_PA

Resumen de: US2022300800A1

Various embodiments set forth systems and techniques for adaptive generation and visualization of a quantized neural network. The techniques include extracting, based on one or more input features and one or more non-quantized network parameters, one or more attributes; calculating, based on the one or more attributes, one or more quantization coefficients; generating, based on the one or more quantization coefficients, one or more quantized input features; and generating, based on the one or more quantized input features and one or more quantization techniques, a neural network.

traducir

Visual Model for Image Analysis of Material Characterization and Analysis Method Thereof

NºPublicación: US2022301139A1 22/09/2022

Solicitante:

CITIC DICASTAL CO LTD [CN]

CN_113177574_A

Resumen de: US2022301139A1

The present disclosure provides a visual model for image analysis of material characterization and analysis method thereof. By collecting and labeling big data of microscopic images, the present disclosure establishes an image data set of material characterization; and uses this data set for high-throughput deep learning, establishes a neural network model and dynamic statistical model based on deep learning, to identify and locate atomic or lattice defects, and automatically mark the lattice spacing, obtain the classification and statistics of the true shape of the microscopic particles of the material, quantitatively analyze the tissue dynamics of the material.

traducir

SYSTEMS AND METHODS FOR IDENTIFYING ITEM SUBSTITUTIONS

NºPublicación: US2022301037A1 22/09/2022

Solicitante:

THE BOSTON CONSULTING GROUP INC [US]

Resumen de: US2022301037A1

Systems and methods for identifying item substitutions. History information can be collected. The history information can include one or more episodes from one or more customers. Each episode can include one or more items. The history information can be transformed into a matrix of observed substitutions. A neural network can be trained on the matrix of observed substitutions to generate item embeddings. Input including an item can be received. A substitution similarity between the item and another item based on the item embeddings can be identified. The item embeddings can be clustered to generate need states.

traducir

NEURAL NETWORK AND CLASSIFIER SELECTION SYSTEMS AND METHODS

NºPublicación: US2022301274A1 22/09/2022

Solicitante:

FLIR COMM SYS INC [US]

CN_113874877_PA

Resumen de: US2022301274A1

High resolution image target classification systems and methods include a proposal component configured to receive a first set data associated with a scene, the first set of data including at least one image of the scene, a multistage neural network comprising a plurality of neural networks, each neural network trained to receive a region of interest and output an object classification in accordance with an associated resource allocation, and an attention coordinator configured to determine regions of interest in the image and allocate each determined region to one of the plurality of neural networks from the multi-scale neural network, in accordance with available system resources. The system may be configured to optimize a probability of detecting objects in the image, while minimizing a number of pixels processed through the multi-scale neural network.

traducir

TECHNIQUES FOR ADAPTIVE GENERATION AND VISUALIZATION OF QUANTIZED NEURAL NETWORKS

NºPublicación: WO2022197615A1 22/09/2022

Solicitante:

VIANAI SYSTEMS INC [US]

US_2022300800_PA

Resumen de: WO2022197615A1

Various embodiments set forth systems and techniques for adaptive generation and visualization of a quantized neural network. The techniques include extracting, based on one or more input features and one or more non-quantized network parameters, one or more attributes; calculating, based on the one or more attributes, one or more quantization coefficients; generating, based on the one or more quantization coefficients, one or more quantized input features; and generating, based on the one or more quantized input features and one or more quantization techniques, a neural network.

traducir

METHOD AND DEVICE FOR PRESENTING PROMPT INFORMATION AND STORAGE MEDIUM

Nº publicación: EP4060548A1 21/09/2022

Solicitante:

FUJITSU LTD [JP]

CN_115114919_PA

Resumen de: EP4060548A1

Method and device for presenting prompt information and a storage medium are provided. The method includes: inputting to the neural network an electronic text including an entity and a context of the entity, a type of the entity, a part of speech of the context, and multiple predefined concepts in text form, wherein the neural network includes a BERT model and a graph convolutional neural network; generating a first vector based on a combination of the entity, the context, the type of the entity and the part of speech of the context by using the BERT model; generating a second vector based on each of the multiple concepts by using the BERT model; generating a third vector based on a graph by using the graph convolutional neural network, wherein the graph is generated based on the multiple concepts and relationships among the concepts; generating a fourth vector by concatenating the second vector and the third vector; calculating a semantic similarity between the entity and each of the multiple concepts based on the first vector and the fourth vector; determining, based on the first vector and the semantic similarity, that the entity corresponds to one of the multiple concepts; and generating the prompt information to be presented to the user based on the determined concept.

traducir

Página1 de 4 nextPage por página

punteroimgVolver