NEURONAL NETWORKS

VolverGo back

Resultados 93 results. LastUpdate Updated on 29/09/2022 [17:45:00] pdf PDF xls XLS

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



previousPage Page2/4 nextPage   results/page


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

Publication No.: EP4060367A1 21/09/2022

Applicant:

FRAUNHOFER GES FORSCHUNG [DE]

WO_2022194516_PA

Absstract of: EP4060367A1

The present disclosure relates to a concept for estimating 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

Automatic delineation and extraction of tabular data using machine learning

Publication No.: GB2605052A 21/09/2022

Applicant:

IBM [US]

DE_112020005095_T5

Absstract of: GB2605052A

A computer-implemented method for using a machine learning model(122) to automatically extract tabular data from an image includes receiving a set of images of tabular data and a set of markup data corresponding respectively to the images of tabular data. The method further includes training a first neural network to delineate the tabular data into cells(440) using the markup data, and training a second neural network to determine content of the cells(440)in the tabular data using the markup data. The method further includes, upon receiving an input image(112) containing a first tabular data without any markup data, generating an electronic output corresponding to the first tabular data by determining the structure of the first tabular data using the first neural network and extracting content of the first tabular using the second neural network.

traducir

DATA PROCESSING METHOD AND DEVICE FOR TRAINING NEURAL NETWORK THAT CATEGORIZES INTENT IN NATURAL LANGUAGE

Publication No.: WO2022191368A1 15/09/2022

Applicant:

MYDATA LAB CO LTD [KR]

Absstract of: WO2022191368A1

A data processing method performed by a computing device is disclosed. A data processing method for natural language used for training a neural network that categorizes intent in natural language, according to an embodiment, may comprise: a step of receiving first data, which is natural language used for training a neural network that categorizes intent in natural language; and a step of pre-processing the first data on the basis of at least one of a first database, which stores a list of synonyms, and a second database, which stores a list of commonly used words unrelated to categorizing intent in natural language.

traducir

Training Data Generation for Artificial Intelligence-Based Sequencing

Publication No.: US2022292297A1 15/09/2022

Applicant:

ILLUMINA INC [US]

JP_2022535306_A

Absstract of: US2022292297A1

The technology disclosed relates to generating ground truth training data to train a neural network-based template generator for cluster metadata determination task. In particular, it relates to accessing sequencing images, obtaining, from a base caller, a base call classifying each subpixel in the sequencing images as one of four bases (A, C, T, and G), generating a cluster map that identifies clusters as disjointed regions of contiguous subpixels which share a substantially matching base call sequence, determining cluster metadata based on the disjointed regions in the cluster map, and using the cluster metadata to generate the ground truth training data for training the neural network-based template generator for the cluster metadata determination task.

traducir

DYNAMIC GESTURE RECOGNITION METHOD, GESTURE INTERACTION METHOD, AND INTERACTION SYSTEM

Publication No.: WO2022188259A1 15/09/2022

Applicant:

OMNIVISION SENSOR SOLUTION SHANGHAI CO LTD [CN]

CN_112949512_A

Absstract of: WO2022188259A1

Disclosed in the present invention are a dynamic gesture recognition method, a gesture interaction method, and an interaction system. The interaction system comprises: a dynamic vision sensor for triggering an event on the basis of a relative motion between an object in a field of view and the dynamic vision sensor, and outputting an event data stream to a hand detection module; the hand detection module for processing the event data stream to determine an initial hand position; a hand tracking module for determining, on the basis of the initial hand position, a series of state vectors that indicate a hand motion state in the event data stream; a gesture recognition module for constructing an event cloud on the basis of event data pointed by the obtained state vectors, and processing the event cloud by using a point cloud-based neural network to recognize a gesture category; and an instruction response module for executing a corresponding operation instruction on the basis of the recognized gesture category. Also disclosed in the present invention is a corresponding computing device.

traducir

SURVIVAL TIME PREDICTION METHOD AND SYSTEM BASED ON IMAGING GENOMICS

Publication No.: WO2022188490A1 15/09/2022

Applicant:

SHENZHEN INST OF ADV TECH CAS [CN]

CN_112907555_A

Absstract of: WO2022188490A1

Disclosed in the present invention is a survival time prediction method and system based on imaging genomics. The method comprises: obtaining image data of tumor patients and survival time data and RNA data of the patients, to establish a data set; separating tumor regions of the patients from the image data; inputting the image data of the patients into a neural network to extract image features and cluster same to obtain a plurality of image modules; using the RNA data to obtain gene modules of the patients; performing screening according to correlations between the gene modules and the image modules to select a plurality of strongly correlated gene modules and image modules; performing pathway enrichment on genes in the selected gene modules to obtain gene pathways related to the image modules; calculating gene set variation analysis scores of the gene pathways, and retaining a gene pathway strongly correlated to the image modules; and using retained image features to perform survival time prediction. The present invention can improv the biological interpretability of survival time prediction, and also improves the generalization capability of deep learning.

traducir

MULTIPLE-FACTOR RECOGNITION AND VALIDATION FOR SECURITY SYSTEMS

Publication No.: US2022292902A1 15/09/2022

Applicant:

INTELLIVISION TECH CORP [US]

EP_4057167_PA

Absstract of: US2022292902A1

A security system can conditionally grant or deny access to a protected area using an artificial intelligence system to analyze images. In an example, an access control method can include receiving candidate information about a face and gesture from a first individual and receiving other image information from or about a second individual. The candidate information can be analyzed using a neural network-based recognition processor that can provide a first recognition result indicating whether the first individual corresponds to a first enrollee of the security system, and can provide a second recognition result indicating whether the second individual corresponds to a second enrollee of the security system. The example method can include receiving a passcode, such as from the first individual. Access can be conditionally granted or denied based on the passcode and the recognition results.

traducir

NEURAL NETWORK PROCESSING UNIT, NEURAL NETWORK PROCESSING METHOD AND DEVICE

Publication No.: US2022292337A1 15/09/2022

Applicant:

BEIJING BAIDU NETCOM SCI & TECH CO LTD [CN]

JP_2022116266_A

Absstract of: US2022292337A1

A neural network processing method, a neural network processing unit (NPU) and a processing device are provided. The method includes: obtaining by a quantizing unit in the NPU float type input data, quantizing the float type input data to obtain quantized input data, and providing the quantized input data to an operation unit; performing by the operation unit of the NPU a matrix-vector operation and/or a convolution operation to the quantized input data to obtain an operation result of the quantized input data; and performing by the quantizing unit inverse quantization to the operation result output by the operation unit to obtain an inverse quantization result.

traducir

IDENTIFYING TRENDS USING EMBEDDING DRIFT OVER TIME

Publication No.: US2022292340A1 15/09/2022

Applicant:

CAPITAL ONE SERVICES LLC [US]

WO_2022192270_PA

Absstract of: US2022292340A1

Systems, methods, and computer program products for identifying trends in behavior using embedding drift. A graph neural network may receive a network graph includes a plurality of nodes, the network graph based on a plurality of transactions for a first time interval, each transaction associated with at least one account. An embedding layer of the neural network may generate, based on the network graph, a respective embedding vector for each of the nodes. The neural network may receive a second embedding vector for each of the nodes. The neural network may determine, based on the embedding vectors and the second embedding vectors, a respective drift for each node. The neural network may determine that the drift of a first node is greater than the drift of a second node, and performing a processing operation on a first account corresponding to the first node.

traducir

METHOD OF TRAINING A NEURAL NETWORK TO REFLECT EMOTIONAL PERCEPTION AND RELATED SYSTEM AND METHOD FOR CATEGORIZING AND FINDING ASSOCIATED CONTENT

Publication No.: US2022292355A1 15/09/2022

Applicant:

EMOTIONAL PERCEPTION AI LTD [GB]

JP_2022528564_A

Absstract of: US2022292355A1

A property vector representing extractable measurable properties, such as musical properties, of a file is mapped to semantic properties for the file. This is achieved by using artificial neural networks “ANNs” in which weights and biases are trained to align a distance dissimilarity measure in property space for pairwise comparative files back towards a corresponding semantic distance dissimilarity measure in semantic space for those same files. The result is that, once optimised, the ANNs can process any file, parsed with those properties, to identify other files sharing common traits reflective of emotional-perception, thereby rendering a more liable and true-to-life result of similarity/dissimilarity. This contrasts with simply training a neural network to consider extractable measurable properties that, in isolation, do not provide a reliable contextual relationship into the real-world.

traducir

SYSTEMS AND METHODS OF TRAINING NEURAL NETWORKS AGAINST ADVERSARIAL ATTACKS

Publication No.: US2022292356A1 15/09/2022

Applicant:

ADOBE INC [US]

Absstract of: US2022292356A1

Embodiments disclosed herein describe systems, methods, and products that generate trained neural networks that are robust against adversarial attacks. During a training phase, an illustrative computer may iteratively optimize a loss function that may include a penalty for ill-conditioned weight matrices in addition to a penalty for classification errors. Therefore, after the training phase, the trained neural network may include one or more well-conditioned weight matrices. The one or more well-conditioned weight matrices may minimize the effect of perturbations within an adversarial input thereby increasing the accuracy of classification of the adversarial input. By contrast, conventional training approaches may merely reduce the classification errors using backpropagation, and, as a result, any perturbation in an input is prone to generate a large effect on the output.

traducir

AUTOMATIC GENERATION OF STATEMENT-RESPONSE SETS FROM CONVERSATIONAL TEXT USING NATURAL LANGUAGE PROCESSING

Publication No.: US2022292260A1 15/09/2022

Applicant:

SCORPCAST LLC [US]

US_2022012428_A1

Absstract of: US2022292260A1

Systems and methods that access an online networked resource using a locator are disclosed. A first item of content on the networked resource is identified. A trigger rule comprising keywords and a sentiment classifier is accessed. A neural network, including input, hidden, and output layers, is used to assign a sentiment classification to the first item of content. The trigger rule, the sentiment classification, and identified keywords, are used to determine whether response content is to be posted to the online networked resource. In response to determining, using the trigger rule, the assigned sentiment classification, and keywords identified in the first item of content, that response content is to be posted to the online networked resource, the sentiment classification and identified keywords are used to select and/or generate a second item of content, and the second item of content is enabled to be posted to the online networked resource.

traducir

USER AUTHENTICATION USING ORIGINAL AND MODIFIED IMAGES

Publication No.: US2022292171A1 15/09/2022

Applicant:

IBM [US]

WO_2022193862_PA

Absstract of: US2022292171A1

Method and system are provided for user authentication using original and modified images. The method includes receiving an original image for a user, where the original image is a private image that meets certain configured criteria. The method uses a pre-trained Convolutional Neural Network (CNN) model to extract one or more image features of the original image and feeds the extracted image feature into a Generative Adversarial Network (GAN) image generator to realistically modify the extracted image feature to generate a modified image. The method authenticates a user based on recognition of a presented original image or a presented modified image.

traducir

SYSTEM AND METHOD FOR CLUSTERING EMAILS IDENTIFIED AS SPAM

Publication No.: US2022294751A1 15/09/2022

Applicant:

AO KASPERSKY LAB [RU]

RU_2769633_C1

Absstract of: US2022294751A1

Disclosed herein are systems and methods for clustering email messages identified as spam using a trained classifier. In one aspect, an exemplary method comprises, selecting at least two characteristics from each received email message, for each received email message, using a classifier containing a neural network, determining whether or not the email message is a spam based on the at least two characteristics of the email message, for each email message determined as being a spam email, calculating a feature vector, the feature vector being calculated at a final hidden layer of the neural network, and generating one or more clusters of the email messages identified as spam based on similarities of the feature vectors calculated at the final hidden layer of the neural network.

traducir

METHOD FOR OBJECT DETECTION AND RECOGNITION BASED ON NEURAL NETWORK

Publication No.: US2022292311A1 15/09/2022

Applicant:

DIBI CHONGQING INTELLIGENT TECH RESEARCH INSTITUTE CO LTD [CN]

CN_112884064_A

Absstract of: US2022292311A1

The present disclosure provides a method for object detection and recognition based on a neural network. The method includes: adding a detection layer following three detection layers of an existing YOLOv5 network model, to construct a new YOLOv5 network model; then, training the new YOLOv5 network model by considering an overlapping area between a predicted box and a ground truth box, a center-to-center distance between the two boxes, and an aspect ratio of the two boxes; and finally, inputting a to-be-detected image into the trained new YOLOv5 network model, outputting a predicted box of an object and probability values corresponding to a class to which the object belongs, and setting a class corresponding to a maximum probability value as a predicted class of the object in the to-be-detected image. This method can quickly and effectively detect multiple classes of objects. Especially, a detection effect for small objects is more ideal.

traducir

BIPOLAR MORPHOLOGICAL NEURAL NETWORKS

Publication No.: US2022292312A1 15/09/2022

Applicant:

SMART ENGINES SERVICE LLC [RU]

Absstract of: US2022292312A1

A bipolar morphological neural network may be generated by converting an initial neural network by replacing multiplication calculations in one or more convolutional layers with approximations that utilize maximum/minimum and/or addition/subtraction operations. The remaining part of the network may be trained after each convolutional layer is converted.

traducir

Methods for Compressing a Neural Network

Publication No.: US2022292376A1 15/09/2022

Applicant:

VOLKSWAGEN AG [DE]

CN_114287008_PA

Absstract of: US2022292376A1

The disclosure relates to methods for compressing a neural network, wherein members of a vehicle fleet locally execute the neural network and during at least one inference phase each determine a selection of elements of the neural network that should be pruned, wherein the members of the fleet transmit the respective determined selection to a central server, wherein the central server merges the respective transmitted selections and generates a merged selection, and wherein the central server prunes the neural network on the basis of the merged selection.

traducir

GENERATING REFINED SEGMENTATIONS MASKS VIA METICULOUS OBJECT SEGMENTATION

Publication No.: US2022292684A1 15/09/2022

Applicant:

ADOBE INC [US]

Absstract of: US2022292684A1

The present disclosure relates to systems, methods, and non-transitory computer-readable media that utilizes a neural network having a hierarchy of hierarchical point-wise refining blocks to generate refined segmentation masks for high-resolution digital visual media items. For example, in one or more embodiments, the disclosed systems utilize a segmentation refinement neural network having an encoder and a recursive decoder to generate the refined segmentation masks. The recursive decoder includes a deconvolution branch for generating feature maps and a refinement branch for generating and refining segmentation masks. In particular, in some cases, the refinement branch includes a hierarchy of hierarchical point-wise refining blocks that recursively refine a segmentation mask generated for a digital visual media item. In some cases, the disclosed systems utilize a segmentation refinement neural network that includes a low-resolution network and a high-resolution network, each including an encoder and a recursive decoder, to generate the refined segmentation masks.

traducir

KEY GENERATION APPARATUS AND METHOD BASED ON MACHINE LEARNING

Publication No.: US2022294620A1 15/09/2022

Applicant:

ELECTRONICS & TELECOMMUNICATIONS RES INST [KR]

Absstract of: US2022294620A1

Disclosed herein are a key generation apparatus and method based on machine learning. The key generation method includes generating, by first and second key generation apparatuses, first and second commit values, and uploading the first commit value and the second commit value to an external repository, training, by the first and second key generation apparatuses, a neural network so as to match weight values with each other, sharing, by the first and second key generation apparatuses, the first and second commit values with each other, comparing shared first and second commit values with uploaded commit values, and then verifying the commit values, and when verification of the commit values has succeeded, generating, by the first and second key generation apparatuses, hash values using the matched weight value, verifying whether the hash values are identical to each other, and generating a session secret key based on a result of verification.

traducir

DEEP-LEARNING BASED DEVICE AND METHOD FOR DETECTING SOURCE-CODE VULNERABILITY WITH IMPROVED ROBUSTNESS

Publication No.: US2022292200A1 15/09/2022

Applicant:

UNIV HUAZHONG SCIENCE TECH [CN]

CN_112989358_A

Absstract of: US2022292200A1

The present invention relates a device for improving robustness of deep-learning based detection of source-code vulnerability, the device at least comprises a code-converting module, a mapping module, and a neural-network module, wherein the mapping module is in data connection with the code-converting module, the mapping module is in data connection with the neural-network module, respectively, and the neural-network module includes at least two first classifiers, based on a received first training program source code, the mapping module maps a plurality of code snippets, and the neural-network module trains the at least two first classifiers according to a first sample vector. The present invention improves the robustness of detection of source-code vulnerability by performing classification training on the feature generators and the classifiers.

traducir

EFFICIENT TEST-TIME ADAPTATION FOR IMPROVED TEMPORAL CONSISTENCY IN VIDEO PROCESSING

Publication No.: US2022292302A1 15/09/2022

Applicant:

QUALCOMM INC [US]

WO_2022192449_PA

Absstract of: US2022292302A1

A method for processing a video includes receiving a video as an input at a first layer of an artificial neural network (ANN). A first frame of the video is processed to generate a first label. Thereafter, the artificial neural network is updated based on the first label. The updating is performed while concurrently processing a second frame of the video. In doing so, the temporal inconsistency between labels is reduced.

traducir

System and a Method for Training a Neural Network Having Autoencoder Architecture to Recover Missing Data

Publication No.: US2022292301A1 15/09/2022

Applicant:

MITSUBISHI ELECTRIC RES LABORATORIES INC [US]

Absstract of: US2022292301A1

A computer-implemented method of training an autoencoder to recover missing data is provided. The autoencoder includes an encoder for encoding its inputs into a latent space and a decoder for decoding the encodings from the latent space. The method comprises creating a first training set including a valid data set of multiple dimensions, and training the encoder and the decoder in a first training stage using the first training set to reduce a difference between the valid data set provided to the encoder and a data set decoded by the decoder. The method further comprises creating a second training set comprising an invalid data set, and training the encoder in a second training stage using the second training set to reduce a difference between encodings of valid data instances and encodings of their corresponding invalid data instances.

traducir

SYSTEMS AND METHODS FOR DETECTING WIND TURBINE OPERATION ANOMALY USING DEEP LEARNING

Publication No.: US2022292666A1 15/09/2022

Applicant:

GEN ELECTRIC [US]

CN_114450646_PA

Absstract of: US2022292666A1

A system and method including receiving historical time series sensor data associated with operation of an industrial asset; generating visual representation images of scatter plots based on the historical time series sensor data based on a reference to a digitized knowledge domain associated with the industrial asset; assigning a root cause label to each image; generating a convolutional neural network (CNN) model trained and tested using subsets of the labeled images; and processing, by the CNN model, a real-time image to detect at least one anomaly in the real-time image and one or more root causes associated with the at least one anomaly.

traducir

SYSTEMS AND METHODS FOR DEPTH ESTIMATION IN A VEHICLE

Publication No.: US2022292289A1 15/09/2022

Applicant:

GM GLOBAL TECH OPERATIONS LLC [US]

CN_115082874_PA

Absstract of: US2022292289A1

Methods and system for training a neural network for depth estimation in a vehicle. The methods and systems receive respective training image data from at least two cameras. Fields of view of adjacent cameras of the at least two cameras partially overlap. The respective training image data is processed through a neural network providing depth data and semantic segmentation data as outputs. The neural network is trained based on a loss function. The loss function combines a plurality of loss terms including at least a semantic segmentation loss term and a panoramic loss term. The panoramic loss term includes a similarity measure regarding overlapping image patches of the respective image data that each correspond to a region of overlapping fields of view of the adjacent cameras. The semantic segmentation loss term quantifies a difference between ground truth semantic segmentation data and the semantic segmentation data output from the neural network.

traducir

EFFICIENT QUANTIZATION FOR NEURAL NETWORK DEPLOYMENT AND EXECUTION

Nº publicación: US2022292300A1 15/09/2022

Applicant:

CYPRESS SEMICONDUCTOR CORP [US]

CN_115080139_PA

Absstract of: US2022292300A1

Implementations disclosed describe methods and systems to perform the methods of deploying and executing machine learning models on target-specific computational platforms. Optimization techniques include but are not limited to alignment of kernel operations with hardware instructions of a target processing device, reduction of kernel dimensions near boundaries of data, efficient reuse of a small number of memory components during neural network operations, run-time quantization of data and neural network parameters, and other methods.

traducir

previousPage Page2/4 nextPage results/page

punteroimgGo back