Machine learning

MACHINE LEARNING-BASED METHODS AND SYSTEMS FOR MODELING USER-SPECIFIC, ACTIVITY SPECIFIC ENGAGEMENT PREDICTING SCORES

Resumen de: US2024338604A1

A machine-learning based method includes receiving an instruction to model an engagement predicting score for a user. User-specific, activity-specific data is obtained from digital resources that include a user-specific activity performance data regarding performance of at least one activity by the user, an object data for an object that allows the user to perform the at least one activity, and user-specific personal data of the user. A user-specific activity engagement labeling data for the at least one activity is predicted by utilizing a first-type data pipeline on the at least one user-specific activity performance data. User-specific, activity-specific data features are predicted by utilizing a second-type data pipeline on the user-specific, activity-specific data. The engagement predicting score is predicted from the user-specific, activity-specific data features and the user-specific activity engagement labeling data. A computing device is instructed to present at least one user-specific activity-related action instruction.

LEARNING SYSTEM OF MACHINE LEARNING MODEL FOR PREDICTION OF PEDESTRIAN TRAFFIC

Resumen de: US2024338610A1

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.

Automated Model Generation Platform for Recursive Model Building

Resumen de: US2024338606A1

Aspects of the disclosure relate to an automated model generation platform for recursive model building. A computing platform may receive a request for automated machine learning model building, and may identify a service offering corresponding to the request. Based on the identified service offering and using machine learning algorithms, the computing platform may select machine learning models and a corresponding sequence of model application (e.g., machine learning model information). The computing platform may store the machine learning model information along with a corresponding indication of the identified service offering. The computing platform may receive a request for model information corresponding to a service access request, and may identify that the service access request corresponds to a problem within the identified service offering. In response, the computing platform may send the machine learning model information, which may cause the enterprise service host system to generate a service output interface.

WIND TURBINE CONTROL SYSTEM INCLUDING AN ARTIFICAL INTELLIGENCE ENSEMBLE ENGINE

Resumen de: US2024337249A1

A system for generating power includes an environmental engine that determines performance metrics for a plurality of wind turbines deployed at a plurality of windfarms, such that each windfarm includes a corresponding subset of the plurality of windfarms. The performance metrics for a given wind turbine of the plurality of wind turbines characterizes wind flowing over blades of the given wind turbine. The system includes an artificial intelligence (AI) ensemble engine operating on the one or more computing devices that generates a set of models for each wind turbine of the plurality of wind turbines, wherein each model of each set of models is generated with a different machine learning algorithm and selects, for each respective set of models, a model with a highest efficiency metric. The AI engine provides edge computing systems operating at the plurality of windfarms with a selected model and corresponding recommended operating parameters.

METHOD AND SYSTEM FOR RECOMMENDING A MACHINE LEARNING PIPELINE FOR AN INDUSTRIAL USE CASE

Resumen de: WO2024200467A1

The recommender system uses as input a new use case description (UCD), which is a free-form text description of a new industrial use case for machine learning. A set of available use cases is initialized by filling it with previous industrial use cases for machine learning. A graph database provides a knowledge graph (KG) containing for each available use case at least one attribute, and a machine learning pipeline that has been suitable. A language model (LM) computes a vector space embedding for the new use case description and for available use case descriptions. Relevance scores are computed for each available use case in the embedding space. A questionnaire component (QC) iteratively selects questions, which are binary questions corresponding to an attribute in the knowledge graph that splits the set of available use cases into two sets, wherein the sums of the relevance scores of the available use cases in the two sets are approximately the same, and incrementally filters the available use cases in the knowledge graph by removing use cases from the set of available use cases based on the answer to the question. If there is only one available use case remaining or if a user interface (UI) detects a selection of one of the available use cases, the machine learning pipeline (MLP) that is linked in the knowledge graph to that use case is recommended to a user (U).

REAL-TIME INTRADIALYTIC HYPOTENSION PREDICTION

Resumen de: US2024331866A1

Techniques for real-time intradialytic hypotension (IDH) prediction are disclosed. A system obtains historical hemodialysis treatment data that is segmented into sets of machine learning training data based on temporal proximities to IDH events and trains a machine learning model to predict IDH events based on the sets of machine learning training data.

AUTOMATED DATA SHARING AND ANALYTICS USING A PRIVACY-PRESERVING DATA SPACE PLATFORM

Resumen de: US2024330497A1

A computer-implemented method for performing automated sharing of data and analytics across a data space platform includes receiving a request for a data analytics service from a first data stakeholder and providing an initial analysis to the first data stakeholder based on determining a portion of semantic data of the data space platform that is accessible to the first data stakeholder. The initial analysis is updated based on comparing the portion of semantic data with another portion of semantic data of the data space platform that is accessible to a second data stakeholder. The updated analysis is provided to the first data stakeholder. The method can be applied to machine learning and regression problems (continuous values) including, but not limited to, providing improvements to various technical fields such as medical diagnosis and treatment, operation system design and optimization, material design and optimization, telecommunication network design and optimization.

System, Method, and Computer Program Product for Time-Based Ensemble Learning Using Supervised and Unsupervised Machine Learning Models

Resumen de: US2024330781A1

Provided are systems for ensemble learning with machine learning models that include a processor to receive a training dataset of a plurality of data instances, wherein each data instance comprises a time series of data points, add an amount of time delay to one or more data instances to provide an augmented training dataset, select a first plurality of supervised machine learning models, select a second plurality of unsupervised machine learning models, train the first plurality of supervised machine learning models and the second plurality of unsupervised machine learning models based on the augmented training dataset, generate an ensemble machine learning model based on outputs of the supervised machine learning models and unsupervised machine learning models, and generate a runtime output of the ensemble machine learning model based on a runtime input to the ensemble machine learning model. Methods and computer program products are also provided.

ARTIFICIAL INTELLIGENCE ADVERSARY RED TEAM

Resumen de: US2024333763A1

An AI adversary red team configured to pentest email and/or network defenses implemented by a cyber threat defense system used to protect an organization and all its entities. AI model(s) trained with machine learning on contextual knowledge of the organization and configured to identify data points from the contextual knowledge including language-based data, email/network connectivity and behavior pattern data, and historic knowledgebase data. The trained AI models cooperate with an AI classifier in producing specific organization-based classifiers for the AI classifier. A phishing email generator generates automated phishing emails to pentest the defense systems, where the phishing email generator cooperates with the AI models to customize the automated phishing emails based on the identified data points of the organization and its entities. The customized phishing emails are then used to initiate one or more specific attacks on one or more specific users associated with the organization and its entities.

GENERATING HEURISTICS USING WEAKLY-SUPERVISED MACHINE LEARNING AND DIGITAL TWIN SIMULATIONS

Resumen de: WO2024201122A1

A computer-implemented method for generating and/or adjusting a heuristic function for a machine learning prediction. A machine learning prediction task including a target entity and attribute is received. Semantic relations are explored to generate relevant entities and attributes related to the target entity and attribute. The heuristic function is generated and/or adjusted based on the relevant entities and attributes.

SYSTEM AND METHOD FOR CONSTRUCTING TOP-PERFORMING PIPELINES USING HIERARCHICAL CONFIGURATION SPACE

Resumen de: US2024330756A1

A computer-implemented method for developing a hierarchical machine-learning pipeline can include receiving a hierarchy specification, a set of estimators for the root node, and one or more transformer options for each of the transformer nodes. The hierarchy specification provides a configuration of the root node, transformer nodes, and edges interconnecting the root and transformer nodes. A rank can be obtained for each estimator in the root node. A hierarchy pipeline traverser can then traverse a first child layer of the transformer nodes connected to the root node via one of the edges. A first ranked list of pathways can be determined with respect to the one or more transformer options selected for the first child layer and at least one selected estimator of the root node.

GRAPH MACHINE LEARNING FOR CASE SIMILARITY

Resumen de: US2024330130A1

Herein is machine learning for anomalous graph detection based on graph embedding, shuffling, comparison, and unsupervised training techniques that can characterize an unfamiliar graph. In an embodiment, a computer obtains many known vectors that respectively represent known graphs. A new vector is generated that represents a new graph that contains multiple vertices. The new vector may contain an arithmetic aggregation of vertex vectors that respectively represent multiple vertices and/or a vector that represents a virtual vertex that is connected to the multiple vertices by respective virtual edges. In the many known vectors, some similar vectors that are similar to the new vector are identified. The new graph is automatically characterized based on a subset of the known graphs that the similar vectors represent.

DEVICES AND COMPONENTS FOR MACHINE LEARNING-BASED SIGNAL ERROR CORRECTION AND METHODS OF USE THEREOF

Resumen de: US2024330102A1

The present disclosure enables signal error correction using a first processor and a memory on a first substrate, where the first processor is operationally connected to a second processor on a second substrate and the memory stores computer code having a machine learning model. The first processor executes computer code to: automatically receive from the second processor, a first output signal intended to be received by a target recipient device. The first processor automatically inputs the first output signal into the machine learning model, where the machine learning model determines that the first output signal includes an error signal that would cause a malfunction in the target recipient device, and output an instruction to cause the first processor to generate a second output signal that corrects the error signal. The first processor automatically generates the second output signal and transmits the second output signal to the target recipient device.

SYSTEMS AND METHODS FOR ADVANCED QUERY GENERATION

Resumen de: US2024330281A1

Systems and methods for determining a query for a data store are described. A natural language text may be analyzed using heuristic processing and one or more machine learning models. Query parameters may be determined from the heuristic processing and machine learning and combined to form a query in a query language. In the heuristic processing, parsing rules may be used to remove conditional terms to generate a base question. The base question may be input to the one or more machine learning model to generate a base query which may be combined with query parameters related to the conditional terms.

INERTIAL SENSING OF TONGUE GESTURES

Resumen de: US2024329751A1

This document relates to employing tongue gestures to control a computing device, and training machine learning models to detect tongue gestures. One example relates to a method or technique that can include receiving one or more motion signals from an inertial sensor. The method or technique can also include detecting a tongue gesture based at least on the one or more motion signals, and outputting the tongue gesture.

DELIVERY TIME ESTIMATION USING ATTRIBUTE-BASED PREDICTION OF DIFFERENCE BETWEEN ARRIVAL TIME AND DELIVERY TIME

Resumen de: WO2024206001A1

An online concierge system receives, from a client device associated with a user of the online concierge system, order data associated with an order placed with the online concierge system, in which the order data describes a delivery location for the order. The online concierge system receives information describing a set of attributes associated with the delivery location and accesses a machine learning model trained to predict a difference between an arrival time and a delivery time for the delivery location. The online concierge system applies the model to the set of attributes associated with the delivery location to predict the difference between the arrival time and the delivery time for the delivery location and determines an estimated delivery time for the order based at least in part on the predicted difference. The online concierge system sends the estimated delivery time for the order for display to the client device.

DEEP LEARNING METHODS FOR BIOSYNTHETIC GENE CLUSTER DISCOVERY

Resumen de: AU2022397403A1

The present disclosure relates to computer-implemented methods and systems for identifying biosynthetic gene clusters (BGCs) that encode pathways for the production of secondary metabolites. Secondary metabolites that target genes or gene products that are homologous to, e.g., human genes or gene products may have utility as potential drug compounds.

METHOD AND SYSTEM FOR RECOMMENDING A MACHINE LEARNING PIPE-LINE FOR AN INDUSTRIAL USE CASE

Resumen de: EP4439401A1

The recommender system uses as input a new use case description (UCD), which is a free-form text description of a new industrial use case for machine learning. A set of available use cases is initialized by filling it with previous industrial use cases for machine learning. A graph database provides a knowledge graph (KG) containing for each available use case at least one attribute, and a machine learning pipeline that has been suitable. A language model (LM) computes a vector space embedding for the new use case description and for available use case descriptions. Relevance scores are computed for each available use case in the embedding space. A questionnaire component (QC) iteratively selects questions, which are binary questions corresponding to an attribute in the knowledge graph that splits the set of available use cases into two sets, wherein the sums of the relevance scores of the available use cases in the two sets are approximately the same, and incrementally filters the available use cases in the knowledge graph by removing use cases from the set of available use cases based on the answer to the question. If there is only one available use case remaining or if a user interface (UI) detects a selection of one of the available use cases, the machine learning pipeline (MLP) that is linked in the knowledge graph to that use case is recommended to a user (U).

GENERALIZABLE MACHINE LEARNING MEDICAL PROTOCOL RECOMMENDATION

Resumen de: CN118284896A

Architecture and techniques for providing promotable machine learning recommendations in conjunction with a medical protocol, such as a radiology protocol. In response to receiving a medical examination order request in a standardized input format, the system may output a recommended protocol in accordance with a standardized output format based on machine learning techniques. The system may then execute a mapping procedure that maps site-specific data to the standardized input format and the standardized output format. The site-specific data may include information specific to an entity providing the medical examination order request.

SYSTEM AND METHOD FOR TRAINING MACHINE LEARNING CLASSIFIER

Resumen de: EP4439403A2

Systems and methods are provided for generating training data for a machine-learning classifier. A knowledge representation synthesized based on an object of interest is used to assign labels to content items. The labeled content items can be used as training data for training a machine learning classifier. The labeled content items can also be used as validation data for the classifier.

ENHANCED DEVICE CLASSIFICATION INCLUDING CROWDSOURCED CLASSIFICATIONS FOR INCREASED ACCURACY

Resumen de: US2024323093A1

Systems, methods, and related technologies for classifying a device on a network are described. A method includes capturing device information corresponding to a device on a network. The method inputs unstructured crowdsourced data on the network into a machine learning model to produce structured crowdsourced data. The method classifies the device based on evaluating the device information with the structured crowdsourced data.

FACILITATING TIME ZONE PREDICTION BASED ON ELECTRONIC COMMUNICATION DATA

Resumen de: US2024323156A1

Methods and systems are provided for facilitating time zone prediction using electronic communication data. Electronic message data associated with a message recipient of electronic communications is obtained. The electronic message data includes message delivery data associated with an electronic message and message response data associated with a response, by the message recipient, to a received electronic message. Using a machine learning model and based on the message delivery data and the message response data, a time-zone score is determined for a time zone. Such a time-zone score can indicate a probability the time zone corresponds with the message recipient. Based on the time-zone score, the time zone is identified as corresponding with the message recipient.

Task Completion Path Generator for Instructional Content

Resumen de: US2024321129A1

A computer implemented method includes accessing instructional content that describes a task for completion by a user. Actions described in the instructional content are derived from the instructional content. Telemetry containing logged actions taken by users is accessed and used to identify actions taken that are associated with the task. A machine learning model is used to identify a task completion path endpoint for the instructional content based on the derived actions and actions taken associated with the task.

AUTOMATED FEATURE ENGINEERING

Resumen de: US2024320565A1

Feature engineering, for example, in automated machine learning, can include receiving streaming data representing at least one attribute detected by a sensor over time. Long term point statistics associated with the streaming data can be computed. The streaming data can be quantized into intervals of time windows and short term point statistics based on the intervals can be computed. The long term point statistics and the short term point statistics can be normalized. Dynamic time warping can be applied across the normalized long term point statistics and short term point statistics. A pair of probability distributions can be generated associated with the dynamic time warped normalized long term point statistics and short term point statistics. Based on distance between the mean values of the probability distributions, machine learning input features can be produced. The machine learning input features can be fed to train a machine learning model for detecting anomaly.

UTILIZING MACHINE-LEARNING MODELS TO GENERATE IDENTIFIER EMBEDDINGS AND DETERMINE DIGITAL CONNECTIONS BETWEEN DIGITAL CONTENT ITEMS

Nº publicación: US2024320288A1 26/09/2024

Solicitante:

DROPBOX INC [US]
Dropbox, Inc

Resumen de: US2024320288A1

The present disclosure relates to systems, methods, and non-transitory computer-readable media that utilize machine learning models to generate identifier embeddings from digital content identifiers and then leverage these identifier embeddings to determine digital connections between digital content items. In particular, the disclosed systems can utilize an embedding machine-learning model that comprises a character-level embedding machine-learning model and a word-level embedding machine-learning model. For example, the disclosed systems can combine a character embedding from the character-level embedding machine-learning model and a token embedding from the word-level embedding machine-learning model. The disclosed systems can determine digital connections between the plurality of digital content items by processing these identifier embeddings for a plurality of digital content items utilizing a content management model. Based on the digital connections, the disclosed systems can surface one or more digital content suggestions to a user interface of a client device.