Machine learning

METHODS AND SYSTEMS FOR ENHANCED SENSOR ASSESSMENTS FOR PREDICTING SECONDARY ENDPOINTS

Resumen de: US2024346324A1

A method, system, and computer-program product includes identifying a set of heterogeneous sensors, configuring a plurality of model training compositions for each of the set of heterogeneous sensors, computing, for each of the plurality of model training compositions, a first efficacy metric value based on predictive outputs of the at least two machine learning models, identifying, for each sensor of the set of heterogeneous sensors, a champion model training composition of the subject sensor, the champion model training composition having a highest efficacy metric value, and electing, from a plurality of champion model training compositions corresponding to the champion model training compositions identified for each sensor of the set of heterogeneous sensors, an overall champion model training composition corresponding to a champion sensor of the set of heterogeneous sensors based on an assessment of second efficacy metric values of the plurality of champion model training compositions.

METHODS AND SYSTEMS FOR ENHANCED SENSOR ASSESSMENTS FOR PREDICTING SECONDARY ENDPOINTS

Resumen de: US2024346382A1

A method, system, and computer-program product includes identifying a set of heterogeneous sensors, configuring a plurality of model training compositions for each of the set of heterogeneous sensors, computing, for each of the plurality of model training compositions, a first efficacy metric value based on predictive outputs of the at least two machine learning models, identifying, for each sensor of the set of heterogeneous sensors, a champion model training composition of the subject sensor, the champion model training composition having a highest efficacy metric value, and electing, from a plurality of champion model training compositions corresponding to the champion model training compositions identified for each sensor of the set of heterogeneous sensors, an overall champion model training composition corresponding to a champion sensor of the set of heterogeneous sensors based on an assessment of second efficacy metric values of the plurality of champion model training compositions.

SYSTEM, METHOD, AND COMPUTER-ACCESSIBLE MEDIUM TO VERIFY DATA COMPLIANCE BY ITERATIVE LEARNING

Resumen de: US2024346048A1

An exemplary system, method, and computer-accessible medium can include, for example, establishing a unique rule-identifier in one-to-one correspondence with at least one set of unknown time-variable rules against which data is to be made compliant, obtaining at least one set of data marked compliant against the one or more set of rules, obtaining meta-data from the compliant data, obtaining at least one set of data marked non-compliant against the set of unknown time-variable rules, extracting meta-data from the non-compliant data, joining the set of compliant and non-compliant metadata to generate a set of estimated rules corresponding to the rule-identifier based at least one of (i) the meta-data of the joined set and (ii) machine learning algorithms.

MACHINE LEARNING PREDICTION OF TEXT TO HIGHLIGHT DURING LIVE AUTOMATED TEXT TRANSCRIPTION

Resumen de: US2024346250A1

A method including transcribing, automatically, an ongoing stream of voice data into text phrases. The method also includes receiving an indication of a selected text phrase in the text phrases. The method also includes converting the selected text phrase to a selected phrase vector. The method also includes generating a subsequent text phrase, after the selected text phrase, from the ongoing stream of voice data, and adding the subsequent text phrase to the text phrases. The method also includes converting the subsequent text phrase to a subsequent phrase vector. The method also includes generating a similarity confidence score from the selected phrase vector and the subsequent phrase vector, using a machine learning model. The method also includes highlighting, responsive to the similarity confidence score exceeding a threshold value, the subsequent text phrase in the text phrases.

REPORTING TAXONOMY

Resumen de: US2024346452A1

A method, computer system, and computer program product are provided for managing reports. A subset of data fields is identified for inclusion in a new report. An intent of the new report is determined based on the subset of data fields. The intent is determined using a set of machine learning models trained from a set of existing reports and a taxonomy of human capital management (HCM) information. Based on the intent determined by the artificial intelligence system, a set of additional fields is predicted for the new report. The set of the additional fields is displayed in a graphical user interface on a display system.

EXPLAINABLE CLASSIFICATIONS WITH ABSTENTION USING CLIENT AGNOSTIC MACHINE LEARNING MODELS

Resumen de: US2024346283A1

Embodiments relate to providing explainable classifications with abstention using client agnostic machine learning models. A technique includes classifying, by a processor, a record with a label using a machine learning model, the machine learning model abstaining from classifying a given record in response to the given record being outside of a scope of an information technology (IT) domain. The processor generates an explanation of a decision by the machine learning model to classify the record with the label and displays the explanation in a human readable form.

EXPLAINABLE CLASSIFICATIONS WITH ABSTENTION USING CLIENT AGNOSTIC MACHINE LEARNING MODELS

Resumen de: US2024345905A1

Embodiments relate to providing explainable classifications with abstention using client agnostic machine learning models. A technique includes inputting, by a processor, records to a machine learning model, the records being associated with an information technology (IT) domain. The technique includes classifying, by the processor, the records with labels using the machine learning model, the machine learning model abstaining from classifying a given record in response to the given record being outside of a scope of the IT domain.

ENHANCED VALIDITY MODELING USING MACHINE-LEARNING TECHNIQUES

Resumen de: US2024346347A1

The present disclosure generally relates to a primary load management system configured to execute machine learning and artificial intelligence techniques to generate predictions of access-right requests that are or are likely to be invalid before the access-right requests are processed for assignment to users or user devices. The present disclosure relates to systems and methods that collect a data set representing characteristics of user devices as the user devices interact with various systems of the primary load management system and train a machine-learning model to predict invalid access-right requests using the collected data set. The collected data set may include a log line that represents each user device, and each log line may be labeled based on an invalidity evaluation. New access-right requests can be processed using the trained machine-learning model to determine whether or not to assign access rights in response to the access-right request.

MID-CYCLE PREDICTIONS USING MACHINE LEARNING

Resumen de: US2024346345A1

Implementations are disclosed for training and/or applying a single machine learning model to generate mid-cycle inferences based on variable length time series inputs. In some implementations, a time series of satellite imagery samples that depict an agricultural area over all or part of a crop cycle may be obtained. During training, ground truth agricultural classifications may be obtained for geographic units of the agricultural area represented by individual pixels of the satellite imagery samples. During training, sample(s) of the time series may be masked to generate a partially masked satellite imagery samples, which in turn may be used to generate input embedding(s). The input embedding(s) may be applied across a machine learning model to generate output(s) representing in-season agricultural prediction(s). During training, the in-season agricultural prediction(s) may be compared with the ground truth agricultural classifications to train the machine learning model.

AUTOMATED MACHINE LEARNING SYSTEM

Resumen de: US2024346375A1

An example method may include receiving, via a graphical user interface, credentials for connecting to a data store containing a plurality of datasets and connecting to the data store using the credentials. A selection of a target metric to predict using the machine learning model can be received, via the graphical user interface, and datasets included in the plurality of datasets that correlate to the target metric can be identified by analyzing the datasets to identify an association between the target metric and data contained within the datasets. The datasets can be input to the machine learning model to train the machine learning model to generate predictions of the target metric, and the machine learning model can be deployed to computing resources in a service provider environment to generate predictions associated with the target metric.

SYSTEMS, METHODS AND DEVICES FOR THE IDENTIFICATION OF CONTENT CREATORS USING MACHINE LEARNING

Resumen de: US2024346370A1

Described herein are systems, methods, and devices for scoring digital content creators and their creations. The systems, methods, devices describe herein enable the content buyers to describe a desired work product and find the ideal content creator for the project.

BEAM-BASED MACHINE LEARNING-ENABLED RF FINGERPRINT (RFFP) POSITIONING

Resumen de: CN118355707A

Various aspects presented herein may enable an ML module to associate RF fingerprints with beam directions and/or beam features to improve uniqueness of the RF fingerprints. In one aspect, a network entity may receive a plurality of first RF fingerprints from one or more wireless devices, each first RF fingerprint of the plurality of first RF fingerprints being associated with at least one orientation feature and a location. The network entity may receive a request to determine a location of a UE based on at least one second RF fingerprint associated with or captured by the UE. The network entity may estimate the location of the UE based at least in part on matching the at least one second RF fingerprint with at least one RF fingerprint of the plurality of first RF fingerprints.

A COMPUTER-IMPLEMENTED METHOD FOR A MACHINE LEARNING-BASED SOFTWARE INFRASTRUCTURE

Resumen de: WO2023105432A1

Computer-implemented method for a software infrastructure, wherein the software infrastructure is based on machine learning techniques and is intended for the analysis of an input dataset (1) containing data obtained from analyses carried out in projects and/or in objects of a predetermined type, the method providing the following steps : a) identification of at least one independen t variable (10) within the input dataset, b) identification of at least one dependent variable (11) within the input dataset, c) identification of existing patterns (12) between the independent variable and the dependent variable, step c) providing a step of hyperparameter adjusting (121) of the machine learning algorithm and a training step (122), d) generating an output dataset (13) containing an estimate of the values of the dependent variable. There is also a step of analysing the input dataset in order to calculate a set of metrics (101) that characterize the input dataset. The step of hyperparameter adjusting (121) provides the following steps : selection of a type of reference machine learning algorithm (102), - identification of the combination of the values of the hyperparameters (103) that most likely allows obtaining the best performance of the selected machine learning algorithm.

GENERATING MACHINE-LEARNING MODEL FOR DOCUMENT EXTRACTION

Resumen de: EP4446948A1

Systems and methods and disclosed that relate to electronic document processing and, more specifically, to generating a machine-learning (ML) model for extracting information from one or more electronic documents, where the ML model can be used as a data object, which can be part of a database command or as part of a document information extraction process that is continuously running, e.g., document information extraction pipeline.

NETWORK DEVICES ASSISTED BY MACHINE LEARNING

Resumen de: US2024340295A1

Devices and methods to identify malicious usage of a network device. In at least one embodiment, a network device comprises circuitry for performing a networking function and collecting telemetry data indicative of the performance of the networking function. The network device obtains an inference of a network traffic pattern using a machine learning model, and responds to the inference.

MANAGEMENT OF MACHINE LEARNING ENTITY INFERENCE EMULATION IN A CELLULAR NETWORK

Resumen de: WO2024211680A1

A device, a method, a system and one or more computer-readable media. A first example device is to host a management service (MnS) producer for a wireless cellular network. One or more processors of the first device are to receive, from an MnS consumer, a request to perform AI/ML emulation in one or more available machine learning (ML) emulation environments; and send to the MnS consumer one or more instances of an information object class (IOC) associated with the process of the AI/ML emulation. A second example device is to host an MnS consumer. One or more processors of the second device are to send, to an MnS producer, a request to perform AI/ML emulation in one or more available machine learning (ML) emulation environments; and receive, from the MnS producer one or more instances of an information object class (IOC) associated with the process of the AI/ML emulation.

DEVELOPMENT PLATFORM FOR IMAGE PROCESSING PIPELINES THAT USE MACHINE LEARNING

Resumen de: WO2024210969A1

A computer system implements a machine learning (ML) pipeline on a chip containing a plurality of hardware compute elements. The computer system accesses a functional description of the ML pipeline that specifies a plurality of functional modules that form the pipeline. At least one functional module corresponds to an ML model. The computer system synthesizes the functional description into a plurality of interconnected executable components that are executable on the hardware compute elements of the chip. In particular, the functional modules are synthesized into executable components that are executable by different hardware compute elements of the chip. The computer system then generates an implementation package including the executable components and specifying their interconnections.

PREDICTING OPTIMAL TREATMENT REGIMEN FOR NEOVASCULAR AGE-RELATED MACULAR DEGENERATION (NAMD) PATIENTS USING MACHINE LEARNING

Resumen de: US2024339191A1

A method and system for predicting a selected treatment regimen for a subject. Baseline data for a subject diagnosed with neovascular age-related macular degeneration (nAMD) is received. A plurality of predictor inputs is formed for an outcome predictor using the baseline data and regimen data for a plurality of treatment regimens. The plurality of predictor inputs comprises a different predictor input for each of the plurality of treatment regimens. A plurality of treatment scores is generated for the plurality of treatment regimens via the set of outcome predictor using the plurality of predictor inputs. One of the plurality of treatment regimens is selected as a selected treatment regimen for the subject based on the plurality of treatment scores.

UTILIZING MACHINE LEARNING AND A SMART TRANSACTION CARD TO AUTOMATICALLY IDENTIFY ITEM DATA ASSOCIATED WITH PURCHASED ITEMS

Resumen de: US2024338724A1

A device may receive, from a client device associated with a customer, an agreement of the customer to join a rewards program, and may receive, from the client device and based on the customer joining the rewards program, item data identifying an item placed in a shopping cart by the customer and customer data identifying the customer, wherein the item data is received by a transaction card from a price tag of the item, and wherein the item data is received after the item has been removed from a shelf and placed in the shopping cart. The device may receive rewards data identifying rewards associated with a plurality of items, and may process the item data, the rewards data, and the customer data, with a machine learning model, to identify a reward for the customer. The device may provide, to the client device, data identifying the reward.

TRANSFER LEARNING METHOD FOR A MACHINE LEARNING SYSTEM

Resumen de: US2024338571A1

A transfer learning method for a system including a plurality of existing agents each including a trained machine learning model for modelling a respective existing machine learning scenario, and a new agent. The system includes a database comprising available models for modelling scenarios, including the trained models, existing scenario metadata indicative of existing scenarios, and transfer learning data indicative of parts of the trained models. The method comprises receiving new scenario metadata indicative of a new scenario to be modelled by the new agent, and receiving new scenario training data for training a model of the new scenario. The method also includes querying the database to: select an available model, based on the received data, to model the new scenario; and, select at least some of the transfer learning data, based on the received data, to train the selected model.

GENERATING MACHINE-LEARNING MODEL FOR DOCUMENT EXTRACTION

Resumen de: US2024338577A1

Systems and methods for generating a machine-learning (ML) model for extracting information from one or more electronic documents, where the ML model can be used as a data object, which can be part of a database command or as part of a document information extraction process that is continuously running (e.g., document information extraction pipeline).

TECHNIQUES FOR PROVIDING EXPLANATIONS FOR TEXT CLASSIFICATION

Resumen de: US2024338531A1

A chatbot system is configured to execute code to perform determining, by the chatbot system, a classification result for an utterance and one or more anchors each anchor of the one or more anchors corresponding to one or more anchor words of the utterance. For each anchor of the one or more anchors, one or more synthetic utterances are generated, and one or more classification results for the one or more synthetic utterances are determined. A report is generated by the chatbot system including a representation of a particular anchor of the one or more anchors, the particular anchor corresponding to a highest confidence value among the one or more anchors. The one or more synthetic utterances may be used to generate a new training dataset for training a machine-learning model. The training dataset may be refined according to a threshold confidence values to filter out datasets for training.

AUTOMATED INPUT-DATA MONITORING TO DYNAMICALLY ADAPT MACHINE-LEARNING TECHNIQUES

Resumen de: US2024338612A1

Systems and methods are disclosed for triggering an update to a machine-learning model upon detecting that a distribution of particular (e.g., recently collected) input data set is sufficiently different from a distribution training input data set used to train the model. The distributions may be determined to be sufficiently different when a classifier can identify to which distribution individual data elements belong (e.g., to at least a predetermined degree). An update to the machine-learning model can include morphing weights used by the model and/or retraining the model.

LEARNING SYSTEM OF MACHINE LEARNING MODEL FOR PREDICTION OF STAY LENGTH IN HOSPITAL

Resumen de: US2024338608A1

A method for automated machine learning includes controlling execution of a plurality of instantiations of different automated machine learning frameworks on a machine learning task each as a separate arm in consideration of available computational resources and time budget. During the execution by the separate arms, a plurality of machine learning models are trained and performance scores of the plurality of trained machine learning models are computed such that one or more of the plurality of trained machine learning models are selectable for the machine learning task based on the performance scores. This invention can be used for predicting patient discharge, predictive control in buildings for energy optimization, and so on.

LEARNING SYSTEM OF MACHINE LEARNING MODEL FOR CLASSIFICATION OF SICKNESS

Nº publicación: US2024338607A1 10/10/2024

Solicitante:

NEC CORP [JP]
NEC Corporation

Resumen de: US2024338607A1

A method for automated machine learning includes controlling execution of a plurality of instantiations of different automated machine learning frameworks on a machine learning task each as a separate arm in consideration of available computational resources and time budget. During the execution by the separate arms, a plurality of machine learning models are trained and performance scores of the plurality of trained machine learning models are computed such that one or more of the plurality of trained machine learning models are selectable for the machine learning task based on the performance scores. This invention can be used for predicting patient discharge, predictive control in buildings for energy optimization, and so on.