Ikaskuntza automatikoa

PAVEMENT MANAGEMENT SYSTEM USING SATELLITE DATA AND MACHINE LEARNING

Resumen de: US20260330619A1

0000 Various examples are provided related to pavement management for maintenance and rehabilitation of roadways. In one example, a method includes obtaining satellite imagery including a portion of a roadway; determining a reflective value corresponding to the portion of the roadway; determining a pavement condition of the portion of the roadway based at least in part upon the reflective value; and identifying a pavement management strategy based at least in part upon the pavement condition. The pavement condition can be determined based upon pavement surface and subsurface conditions of the portion of the roadway. The pavement condition can be determined by a machine learning model trained using a combination of measured, synthesized and/or feature generated training data.

AI Generated Creative Content Based on Shared Memories

Resumen de: US20260330272A1

There is provided a system including a computing platform having a hardware processor and a system memory storing a software code and a memory data structure storing memory features for an artificial intelligence interactive character (AIIC). The hardware processor is configured to execute the software code to elicit, using the AIIC, a reminiscence from a user, predict, using a trained machine learning model and the reminiscence, at least one user memory feature, identify, using the memory data structure, one or more of the memory features for the AIIC based on a similarity between the predicted at least one user memory feature and the memory features for the AIIC, determine a mood modifier based on emotion information of the identified one or more of the memory features for the AIIC, and generate a creative composition for performance by the AIIC based on the mood modifier.

AN ENSEMBLE APPROACH TO PREDICTIVE DAMAGE ANALYTICS

Resumen de: US20260330330A1

The present disclosure relates to systems and methods that use predictive analytics modeling, such as to identify risk profiles as it applies to potential damage to assets, persons, or other items. In one implementation, for example, an artificial intelligence/machine learning (AI) model or engine is adapted to be used as a mechanism for achieving predictive analytics from as-yet unknown adverse future events against one or a plurality of items. The AI model is adapted to evaluate a plurality of known factors for one or a set of evaluation items. The AI model is adapted to build an AI training set based on aggregate data to narrow down a plurality of clusters from the aggregate data, use AI on individual items in each cluster, compare the output across clusters to find one or more ranked elements; and identify individual items through the characteristics to provide the training set.

AUTOMATIC WELL LOG RECONSTRUCTION

Resumen de: US20260328091A1

0000 Machine learning techniques for reconstructing a target well log are presented. The techniques include: storing a dictionary that includes a statistical distribution similarity quantification for each common feature of each pair of well logs in a well log data set; for at least one cluster of the target well log, ranking the well logs in the well log data set based on the dictionary, where the ranking is according to a statistical distribution similarity to the target well log; selecting a validation set of well logs and a training set of well logs; iteratively producing a reconstruction model, where each step of the iteration includes training using the training set of well logs and validating using the validation set of well logs; reconstructing a feature in the target well log using reconstructed values for the feature output by the reconstruction model; and providing the reconstructed target well log.

ARTIFICIALLY INTELLIGENT SYSTEM AND METHOD FOR PROGNOSTIC EVALUATION OF AUTOIMMUNE DISORDERS

Resumen de: US20260331289A1

An artificially intelligent system and method for prognostic evaluation of autoimmune disorders. Certain objects and advantages of the present disclosure include methods and systems configured to aggregate and analyze clinical conversational data, patient biometric data, medical records data, blood biomarker test data, and patient biometric data to tune metrics that are increasingly specific to a patient in the creation of a dynamic patient phenotype to enhance clinical understanding of one or more factors that are specific to the patient's health status. One or more Al framework and engine may facilitate effective integration of communication-related insights into diagnostic and prognostic digital health resources. Exemplary systems, methods, and apparatuses according to the principles herein may comprise machine learning and deep learning techniques for the development of one or more quantitative metrics derived from clinical conversational data.

METHOD AND DEVICE FOR GENERATING SYNTHETIC ANOMALY IMAGES OF A PRODUCT IN ORDER TO ENRICH A TRAINING DATASET

Resumen de: US20260330813A1

A method for generating synthetic anomaly images of a product to enrich a training dataset for anomaly detection and/or anomaly classification of the product. The method includes: providing a user input that describes a product and at least one associated item of anomaly information; extracting a subgraph from a predefined knowledge graph, wherein the knowledge graph includes relationships between products, their properties and/or possible anomalies, and the subgraph refines and/or constrains the information of the user input in a domain-specific manner; generating a plurality of specific text prompts based on the information of the extracted subgraph; using the generated text prompts to control a previously trained text-to-image generation model to generate synthetic anomaly images of the product; and providing the generated anomaly images to enrich a training dataset for training a machine learning model for anomaly detection and/or anomaly classification.

SYSTEM AND METHOD FOR DEEP LEARNING ON ATTACK ENERGY VECTORS

Resumen de: US20260332637A1

A method, computer program product, and computer system for identifying potential social engineering activity associated with two or more communications on two or more communication channels of a plurality of communication channels. Identification may be based on a determination that a current score for the two or more communication channels is greater than a threshold score for a user profile.

COALITION LEARNING FOR TRAINING OF DISTRIBUTED MACHINE LEARNING WORKLOADS

Resumen de: US20260330309A1

A method performed by a network entity in a mobile network is disclosed to determine a configuration for a machine learning workload. The method includes determining that the machine learning workload and one or more other machine learning workloads are members of a same coalition, training the machine learning workload using a first set of trial configurations, sharing a subset of the first set of trial configurations with one or more network entities that are responsible for configuring the one or more other machine learning workloads, obtaining a second set of trial configurations shared by the one or more network entities that are responsible for configuring the one or more other machine learning workloads, and warm starting a configuration optimization process for the machine learning workload using the second set of trial configurations.

AUTOMATED DECISION MAKING FOR MULTIPLE DATASETS WITH MACHINE LEARNING

Resumen de: US20260329916A1

Systems and methods for automated decision making for multiple datasets with machine learning. A decision making entity can be prompted with an interface to generate user queries for entity documents from multiple entities. A query answer can be generated for the entity documents with an iterative question answer controller. The query answer can be converted into a dimensional matrix. Semantic data regarding the query answer can be visualized from the dimensional matrix with a visualization component of the iterative question answer controller to perform downstream tasks to resolve issues with the multiple entities.

SELECTABLE ENCRYPTION FOR 5G OPEN RADIO ACCESS NETWORK

Resumen de: US20260332402A1

0000 Techniques for encrypting data within a communication network include receiving, at a first module, data packets encrypted using mathematical encryption. A machine-learning model trained to detect cybersecurity threats analyzes the data packets to determine the existence of a cybersecurity threat. In response to determining a threat, encryption is switched from mathematical encryption to quantum encryption. The data packets are encrypted using a quantum encryption key to generate quantum-encrypted data packets. The quantum encryption key is transmitted from the first module to a second module over a quantum key distribution (QKD) channel, and the quantum-encrypted data packets are transmitted from the first module to the second module.

ARTIFICIAL INTELLIGENCE SYSTEM PROVIDING INTERACTIVE MODEL INTERPRETATION AND ENHANCEMENT TOOLS

Resumen de: US20260330260A1

An interactive interpretation session with respect to a first version of a machine learning model is initiated. In the session, indications of factors contributing to a prediction decision are provided, as well indications of candidate model enhancement actions. In response to received input, an enhancement action is implemented to obtain a second version of the model. The second version of the model is stored.

ENHANCED VALIDITY MODELING USING MACHINE-LEARNING TECHNIQUES

Resumen de: US20260330262A1

0000 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.

DETECTING UNAUTHORIZED ONLINE APPLICATIONS USING MACHINE LEARNING

Resumen de: US20260330480A1

A method of detecting whether electronic fraud alerts are false positives includes receiving data detailing a financial transaction, inputting the data into a rules-based engine that determines whether to generate an electronic fraud alert for the financial transaction based upon the data, and, when an electronic fraud alert is generated, inputting the data into a machine learning program trained to identify one or more facts indicated by the data. The method may also include determining whether the identified facts can be verified by customer data and, in response to determining that the facts can be verified, retrieving or receiving first customer data. The method may further include verifying that the electronic fraud alert is not a false positive based upon analysis of the first customer data, and transmitting the verified electronic fraud alert to a mobile device of the customer to alert the customer to fraudulent activity.

MACHINE LEARNING BASED PROACTIVE PREDICTION AND INTERVENTION IN INFORMATION TECHNOLOGY INFRASTRUCTURE

Resumen de: US20260330252A1

0000 An apparatus maintains data structures characterizing interactions between users of information technology infrastructure, obtains information characterizing an additional interaction, and applies the obtained information to a first machine learning model, wherein the first machine learning model is trained utilizing at least portions of the data structures to predict a particular one of a plurality of possible outcomes for the additional interaction. Responsive to the prediction of the particular outcome, the apparatus triggers a second machine learning model, different than the first machine learning model, to generate one or more communications for delivery to one or more of the devices, the one or more communications providing at least a portion of at least one interventive interaction targeted at achieving an actual outcome that is different than the predicted outcome, and causes the one or more communications to be transmitted to the one or more devices over at least one network.

SUGGESTING ACTIONS BASED ON MACHINE LEARNING

Resumen de: EP4826622A2

This document describes techniques for suggesting actions based on machine learning. These techniques determine a task that a user desires to perform, and presents a user interface through which to perform the task. To determine this task, the techniques can analyze content displayed on the user device or analyze contexts of the user and user device. With this determined task, the techniques determine an action that may assist the user in performing the task. This action is further determined to be performable through analysis of functionalities of an application, which may or may not be executing or installed on the user device. With some subset of the application's functionalities determined, the techniques presents the subset of functionalities via the user interface. By so doing, the techniques enable a user to complete a task more easily, quickly, or using fewer computing resources.

PREDICTION OF PHARMACOKINETIC PROPERTIES OF CHEMICAL COMPOUNDS

Resumen de: US20260321363A1

0000 Prediction of pharmacokinetic properties of chemical compounds A computer-implemented method of generating a machine-learning model configured to predict the value of one or more pharmacokinetic parameters of a pharmaceutical compound having a chemical structure comprises: receiving training data comprising a plurality of records, each record comprising: data encoding the chemical structure of a pharmaceutical compound; and the values of one or more pharmacokinetic parameters obtained from measurements taken from a human or animal subject after administration of the compound to the human or animal subject; and training the machine-learning model using the training data. Computer-implemented methods of utilizing the generated machine-learning model, particularly in a drug design context, are also provided.

IDENTIFYING AND PROCESSING MARKETING LEADS THAT IMPACT A SELLER'S ENTERPRISE VALUATION

Resumen de: US20260320638A1

0000 Systems, methods and computer readable media for automated marketing decisions based on company valuation results derived from ensemble machine learning models include collecting potential customer data for a set of potential customers from a lead source and transmitting the potential customer data to an optimization system. An ensemble machine learning model that establishes a transaction value for each of the set of potential customers based on an enterprise valuation of the seller. A marketing action is then taken. The marketing action may include one or more of budgeting for a transaction with the lead source, targeting communications to members of the set of potential customers, accepting or rejecting members of the set of potential customers, or offloading members of the set of potential customers.

GENERATING DIGITAL COMPONENTS USING CUSTOMIZED MACHINE LEARNING MODELS

Resumen de: US20260320472A1

0000 One example method includes obtaining, by an artificial intelligence (AI) system, first model parameters of a first machine learning model; generating a second machine learning model based on the first model parameters, the second machine learning model being at a given hierarchical layer of a hierarchical model structure; refining the second machine learning model to obtain an updated second machine learning model comprising second model parameters, wherein the first model parameters and the second model parameters have a different value for at least one model parameter; and generating, by the AI system, a third machine learning model based on the second model parameters, wherein (i) the third machine learning model is at a lower hierarchical layer of the hierarchical model structure beneath the given hierarchical layer of the second machine learning model, and (ii) the second machine learning model and the third machine learning model have a first common attribute.

GENERATING DIGITAL COMPONENTS USING CUSTOMIZED MACHINE LEARNING MODELS

Resumen de: US20260320471A1

One example method includes obtaining, by an artificial intelligence (AI) system, first model parameters associated with a first machine learning model at a first hierarchical layer; generating, by the AI system, a second machine learning model including the first model parameters, the second machine learning model at a second hierarchical layer beneath the first hierarchical layer; and refining, by the AI system, the second machine learning model to obtain an updated second machine learning model including second model parameters, where the first model parameters and the second model parameters have different value for at least one model parameter, and the second model parameters are unshareable with the first machine learning model.

MACHINE-LEARNING A CLASSIFICATION FUNCTION CONFIGURED FOR DETECTION OF MECHANICAL INTERFACES OF A CAD MODEL

Resumen de: US20260320263A1

The disclosure notably relates to a computer-implemented method for machine-learning a classification function configured for detection of mechanical interfaces of a CAD model representing a single mechanical part. The method comprises obtaining a training dataset comprising graphs, each graph representing a CAD model representing a single mechanical part. Each graph comprises one or more nodes representing a canonical surface of the respective CAD model. Each node comprises a label from the list: primary face, secondary face, tertiary face, maintain face, or no interface. Each graph further comprises one or more edges representing a connection between the canonical surfaces. The method further comprises training the classification function based on the dataset to classify, with a label of the label list, each node of a graph representing a CAD model representing a single mechanical part. This provides an improved solution for detecting mechanical interfaces.

METHOD AND SYSTEM FOR MAIZE MAPPING USING MULTI-SOURCE SATELLITE IMAGERY AND MACHINE LEARNING

Resumen de: WO2026216662A1

The present invention relates to a method and system for maize mapping using multi-source satellite imagery and machine learning. The method for maize mapping, according to an embodiment of the present invention, is performed by the system for maize mapping and comprises: a first step of acquiring LiDAR data and optical data captured by a satellite; a second step of training a first random forest classifier by using data of a first area including a height and optical characteristics of maize; a third step of inputting LiDAR data and optical data captured by a satellite of a second area into the trained first random forest classifier to infer maize cultivation areas; a fourth step of filtering outliers in the inferred maize cultivation areas by using phenological growth cycle information of maize; a fifth step of inferring a maize cropping intensity in the filtered maize cultivation areas; and a sixth step of outputting maize location points including point data by excluding paddy field areas from the maize cultivation areas in which the maize cropping intensity is inferred.

MACHINE-LEARNING A CLASSIFICATION FUNCTION CONFIGURED FOR DETECTION OF MECHANICAL INTERFACES OF A CAD MODEL REPRESENTING A MECHANICAL ASSEMBLY OF PARTS

Resumen de: US20260320262A1

0000 A computer-implemented method for machine-learning a classification function configured for detection of mechanical interfaces of a CAD model representing a mechanical assembly of parts. The method includes obtaining a training dataset comprising graphs representing a local contact between parts of a CAD model. The graph includes one or more nodes representing canonical surfaces of the CAD model. Each node comprises a label within the label list: primary face, secondary face, tertiary face, maintain face, and no interface. The graph further includes one or more internal edges and one or more external edges. The method further includes training the classification function based on the training dataset to classify, with a label of the label list, each node of a graph representing a local contact between parts of a CAD model representing a mechanical assembly of parts. This provides an improved solution for detecting mechanical interfaces.

MACHINE LEARNING FOR BINDING AFFINITY PREDICTION USING A SET OF SUPPORT LIGANDS

Resumen de: WO2026217425A1

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for predicting binding affinities of ligands. In one aspect, a method comprises: processing a model input characterizing a test ligand and a target molecule using a global binding affinity computational model to predict a global binding affinity score for the test ligand; selecting a set of support ligands for the test ligand from a database of example ligands; processing a model input that comprises data characterizing the test ligand and each of the support ligands using a local binding affinity machine learning model to generate a predicted error in the global binding affinity score for the test ligand; generating a final binding affinity score for the test ligand by combining the global binding affinity score for the test ligand and predicted error for the test ligand; and outputting the final binding affinity score for the test ligand.

MACHINE UNLEARNING

Resumen de: WO2026217055A1

A computer-implemented method (1000) for machine unlearning is presented The method comprises obtaining (1002) a request (2) indicating one or more X. X represents one or more input samples to be unlearned. The method further comprises for each X: obtaining (1004) U based on X, obtaining (1006) L based on X, updating (1008) F based on W'. W' is obtained so as to increase U, based on W, U, L, LR. F represents a machine learning model trained with at least X. W represents one or more model parameters of F. U represents one or more quantified uncertainty values for X. L represent one or more quantified loss values for X. LR represents a learning rate of F. The method (1000) further comprises initiating (1010) a deploying of F.

SYSTEM AND METHODS FOR POLYMER ANALYSIS

Nº publicación: WO2026215599A1 15/10/2026

Solicitante:

OXFORD NANOPORE TECH PLC [GB]
OXFORD NANOPORE TECHNOLOGIES PLC

Resumen de: WO2026215599A1

A computer-implemented method of processing a measurement signal comprising measurements of a sample by a sensor unit comprising a nanopore includes analysing the measurement signal using a machine learning model to determine that a portion of the measurement signal is associated with a first state of the nanopore, the first state being one of a plurality of states of the nanopore. The method includes determining whether to further analyse the portion of the measurement signal in dependence on the determination that the portion of the measurement signal is associated with the first state of the nanopore.