Alerta

DYNAMIC DATA INGESTION FOR DEPLOYING MACHINE LEARNING MODELS

Resumen de: US20260220541A1

0000 Methods and systems are presented for providing a graph-based framework for evaluating feature candidates for a machine learning model. A production graph is generated to represent relationships among various assets of an organization. The production graph is used by various machine learning models for obtaining input data to perform the corresponding tasks. When it is determined that data corresponding to a feature candidate selected for the machine learning model is missing from the graph, instead of modifying the production graph, a new graph schema that defines one or more additional vertex types or one or more additional edge types is generated. New graph data is also generated based on the new graph schema. In response to a query corresponding to the feature candidate, a merged graph is generated by incorporating the new graph data into the production graph. A query result is obtained based on traversing the merged graph.

SYSTEMS AND METHODS FOR DEEP-AND-WIDE MACHINE LEARNING

Resumen de: WO2026161862A1

Systems and methods for deep-and-wide learning (DWL) in accordance with embodiments of the invention are illustrated. One embodiment includes a system for training a machine learning model, including a processor, and a memory, the memory containing a DWL application that configures the processor to obtain training data includes a plurality of data samples, extract high-dimensional features from the training data, wherein the high-dimensional features capture intra-sample characteristics of individual data samples, extract low-dimensional features from the training data using dimensionality reduction wherein the low-dimensional features capture inter-sample relationships among the plurality of data samples, integrate the high-dimensional features and the low-dimensional features to form a combined feature vector, and train a machine learning model using the combined feature vector.

EFFICIENT ON-DEVICE MACHINE LEARNING MODEL UPDATES

Resumen de: US20260220526A1

Examples relate to systems and methods for providing a machine learning model on a mobile device. The systems and methods store a base machine learning (ML) model on a user system, the base ML model trained to perform a first task in an individual domain. The systems and methods receive input that selects a second task in the individual domain and access parameter update information associated with the second task. The systems and methods update the base ML model based on the parameter update information associated with the second task and generate an output corresponding to the second task by processing an input by the updated base ML model.

ADAPTIVE LOCATION TECHNOLOGY HANDOFFS WITH ASSET TRACKERS

Resumen de: US20260223065A1

Systems and methods are disclosed for determining asset tracker location using multimodal fingerprints and adaptive machine learning techniques. A process obtains data comprising sequences of events, location sources, time values, and sensor measurements to generate multimodal fingerprints. These fingerprints are fused to define “liquid” boundaries with improved certainty. Machine learning models are trained and updated using deep learning architectures that classify and predict asset tracker location based on multimodal fingerprints, motion, and environmental data. Adaptive handoff processes dynamically switch among GPS, Wi-Fi, UWB, dead-reckoning, or other location technologies depending on detected motion changes and operational context, thereby conserving power and optimizing accuracy. An integrated system includes processors, force sensors, radios, inertial sensors, and edge-based AI/ML modules that continuously update location models locally and in conjunction with server-based systems for improved geospatial tracking performance across diverse environments.

System and method for performing a self-supervised training, validation, and execution of an evaluation model for evaluating underperforming machine-learning models

Resumen de: US20260220529A1

0000 A system includes a memory configured to store a set of model parameters associated with a machine-learning model. The machine-learning model is identified as having generated one or more prediction outputs that is deviating from an expected prediction output. The system further includes a processor operably coupled to the memory and configured to, while the machine-learning model is actively generating the one or more prediction outputs, access the set of model parameters, train a first machine-learning model to generate one or more model performance metrics associated with the machine-learning model based on the set of model parameters, in response to training the first machine-learning model, execute the first machine-learning model to generate the one or more model performance metrics, and identify the machine-learning model as underperforming based on whether the one or more model performance metrics satisfies a performance metric threshold.

TELECOMMUNICATION SYSTEM FAILURES PREDICTION THROUGH MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE

Resumen de: US20260222276A1

0000 In an embodiment, a method may be implemented in a computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, the computer system interconnected with a telecommunications system, the method comprising: receiving, at the computer system, data relating to operation of the telecommunication system, obtaining, at the computer system, at least one machine learning model trained to detect and predict faults in the operation of the telecommunication system, selecting, at the computer system, computing infrastructure upon which to execute the at least one machine learning model, wherein the selected computing infrastructure comprises a mesh of interconnected micro-applications, executing, at the computer system, the at least one machine learning model using the selected computing infrastructure to detect and predict faults in the operation of the telecommunication system, and automatically correcting at least some of the detected faults.

HETEROGENEOUS MODEL TREES FOR MACHINE LEARNING WITH DISCONTINUOUS DATASETS

Resumen de: US20260220489A1

A method includes obtaining at least one dataset containing one or more discontinuities, where the one or more discontinuities split data of the at least one dataset into multiple partitions. The method also includes generating feature crosses associated with the at least one dataset. The method further includes generating a decision tree structure based on at least some of the feature crosses, where (i) the decision tree structure includes multiple leaf nodes and (ii) each leaf node corresponds to a different one of the multiple partitions. In addition, the method includes, for each leaf node of the decision tree structure, generating a machine learning model that models data of the corresponding partition.

Machine Learning-Based Information Management Engine

Resumen de: US20260220493A1

A system that trains a machine learning model to determine capture levels for target computations is disclosed. The system utilizes training data encompassing attribute sets and information levels for various computation types. For a form field value computation, the system determines associated attributes. The trained model processes these attributes to establish an appropriate information storage level. Based on this level, the system selects a relevant subset of information related to the computation or its result. The system then stores this selected subset in association with the computed value. The system uses feedback to retrain the model to enhance its performance.

STATUS INFORMATION FOR IDENTIFIERS RELATED TO ARTIFICIAL INTELLIGENCE OR MACHINE LEARNING

Resumen de: US20260222970A1

0000 Methods, systems, and devices for wireless communications are described. A wireless device may obtain, from a network entity, status information corresponding to one or more identifiers that are associated with one or more network settings. The one or more network settings may relate to communication of reference signaling for an artificial intelligence or machine learning (AI/ML)-based positioning or sensing procedure or measurement of reference signaling for an AI/ML-based positioning or sensing procedure. The wireless device may perform an operation to control the AI/ML-based positioning or sensing procedure based on the status information corresponding to the one or more identifiers. The wireless device and the network entity may communicate based on the status information corresponding to the one or more identifiers and the AI/ML-based positioning or sensing procedure control operation.

COACHING OF ARTIFICIAL INTELLIGENCE AGENTS

Resumen de: US20260220543A1

A computer-implemented method for improving an operation of a first adaptive machine-learning model of an artificial intelligence agent by applying a second adaptive machine-learning model with at least one dataset comprising at least one sample is provided, the method comprises: inputting at least one hint as a part of at least one sample in the at least one dataset to the second adaptive machine-learning model; generating a target to an output of the first adaptive machine-learning model of the artificial intelligence agent by applying the at least one hint in an execution of the task with the second adaptive machine-learning model; and training the first adaptive machine-learning model by applying the target generated by the second adaptive machine-learning model. A computing system and a computer program are further provided.

SPECTRAL CLUSTERING FOR SPECTRAL TREE GUIDANCE IN HETEROGENEOUS MODEL TREE GENERATION OR OTHER FUNCTIONS FOR MACHINE LEARNING WITH DISCONTINUOUS DATASETS

Resumen de: US20260220490A1

0000 A method includes obtaining at least one dataset containing one or more discontinuities, where the one or more discontinuities split data of the at least one dataset into multiple partitions. The method also includes performing spectral clustering of the at least one dataset to identify multiple initial clusters of data in the at least one dataset. The method further includes trimming the initial clusters of data in order to identify an estimated number of partitions in the at least one dataset. The method also includes performing spectral clustering of the at least one dataset based on the estimated number of partitions to identify multiple updated clusters of data in the at least one dataset, where each updated cluster of data corresponds to one of the multiple partitions. In addition, the method includes providing the updated clusters of data as input to a machine learning algorithm.

MACHINE LEARNING MODEL EXPLANATION GENERATION

Resumen de: US20260220423A1

In some examples, a system obtains an approximation function for a machine learning (ML) model, the approximation function providing an approximate representation of a latent representation of latent features in a latent space produced by the ML model. The system presents a visualization of the latent features using the approximation function, and the system generates a refined latent space based on user input in the visualization. The system generates an explanation regarding which latent features of the latent space are primary contributors to a decision of the ML model.

SYSTEMS AND METHODS FOR SMART SETTING CONFIGURATION

Resumen de: US20260219896A1

Systems and methods are disclosed for smart setting configurations. Example methods may include ingesting a plurality of items of data, wherein each item of data is retrieved from a data source within a plurality of data sources associated with a user account on a social media platform, for each item of data, providing the item of data as input to a distinct machine learning algorithm in a plurality of contextual machine learning algorithms, and providing output of the plurality of contextual machine learning algorithms to an additional machine learning algorithm as input. Some methods may include formulating a setting configuration for the user account on the social media platform based at least in part on output of the additional machine learning algorithm, and generating a recommendation comprising the setting configuration to a user that operates the user account.

METHODS AND APPARATUS FOR PERSONALIZED GENERATIVE ARTIFICIAL INTELLIGENCE (GENAI) ASSISTANT

Resumen de: WO2026157015A1

An apparatus includes interface circuitry, machine-readable instructions, and at least one processor circuit to be programmed by the machine-readable instructions to obtain a first output from a natural language processing machine learning model engine, the first output generated with retrieval augmented generation (RAG) to respond to a user query, obtain a first feedback answer input associated with the user query, generate an embedding of the first feedback answer input and the user query, cause the embedding to be stored in a vector database to provide feedback for the RAG, and obtain a second output from the natural language processing machine learning model engine, the second output generated in response to the user query, the second output to include the first feedback answer input.

ESTIMATING LOCAL FEATURE IMPORTANCE

Resumen de: US20260220540A1

Here is generation of a local explanation of a machine learning (ML) inference, and this generation is accelerated by accurate estimation of local feature value importances from a multioutput regression by an attribution metamodel. Into a feature vector, a computer stores values for respective features and stores an inference, from the feature values, by an ML model. A multioutput regression by an attribution metamodel includes inferentially generating, from the feature vector, local value importances respectively for the feature values. Based on the local value importances, a local explanation of the inference is generated and displayed. For accelerated training with early stopping, an adaptively selected training corpus involves incrementally adding additional datapoints to the training corpus until the attribution metamodel accurately learns.

SYSTEMS, METHODS, AND APPARATUS FOR AUTOTUNING OF RETRIEVAL AUGMENTED GENERATION PARAMETERS

Resumen de: WO2026157014A1

Systems, apparatus, and methods for autotuning of retrieval augmented generation parameters are disclosed. Example instructions to perform the same include instructions to vectorize a dataset using retrieval augmented generation (RAG) parameters, the RAG parameters identified based on a domain of the dataset, wherein the RAG parameters control at least one of ingestion of data or retrieval of data, execute a machine learning model using the RAG parameters, evaluate a quality of an output of the machine learning model, after determination that the quality does not meet a quality threshold, update the RAG parameters and cause re-execution of the machine learning model using the updated RAG parameters, and after determination that the quality meets the quality threshold, store the RAG parameters in association with the domain of the dataset.

FOUNDATION MODEL DOMAIN AND TASK-SPECIFIC FINE-TUNING

Resumen de: WO2026161237A1

A method includes loading a machine learning model having a set of parameters and trained using initial training data. A new set of training data may be used to determine a set of importance values for the set of petameters indicating a measure of importance for a given parameter to affect an accuracy of an output of the machine learning model. A first plurality of task-specific importance values can be loaded for a particular task using a respective task-specific training set. A second plurality of domain-specific importance values can be loaded for a particular domain using a respective domain-specific training set. Combined importance values can be determined for the set of parameters. The machine learning model can then be further trained using the combined importance values to determine fine-tuned values for the set of parameters.

Agentic approach to multilingual text-to-SQL conversion with intent detection and self-correction

Resumen de: US20260220129A1

0000 An intent associated with a user input is determined. An augmented prompt for generating a Standard Query Language (SQL) code snippet is generated. An SQL code snippet is generated by using the augmented prompt on a machine learning (ML) model. Results are generated by executing the SQL code snippet on a data lake.

REFINED PROMPT - RESPONSE PAIR GENERATION FOR RESPONSE QUALITY MANAGEMENT

Resumen de: US20260220550A1

0000 Methods and systems for managing operation of a request processing pipeline are provided. To service requests, the request processing pipeline may accept unstructured text based requests. The unstructured text may be used to obtain prompts for evaluation by a trained generative machine learning model. The pairs may be evaluated to rank order them with respect to one another. A best ranked one of the pair may be used to service the request. The request may be serviced by providing the response of the best ranked one of the pairs, by initiating provisioning of computer implemented services using the response, and/or via other processes.

SYSTEM AND METHOD FOR DYNAMICALLY PREDICTING FRAUD USING MACHINE LEARNING

Resumen de: US20260220715A1

0000 A computing device configured to communicate with a central server in order to predict likelihood of fraud in current transactions for a target claim. The computing device then extracts from information stored in the central server (relating to the target claim and past transactions for past claims including those marked as fraud), a plurality of distinct sets of features: text-based features derived from the descriptions of communications between the requesting device and the endpoint device, graph-based features derived from information relating to a network of claims and policies connected through shared information, and tabular features derived from the details related to claim information and exposure details. The features are input into a machine learning model for generating a likelihood of fraud in the current transactions and triggering an action based on the likelihood of fraud (e.g. stopping subsequent related transactions to the target claim).

CONSTRAINT BASED INFERENCE AND MACHINE LEARNING SYSTEM

Resumen de: US20260220438A1

0000 A method including receiving a pre-determined constraint on user actions. A constraint vector is generated based on the pre-determined constraint. The constraint vector is input into a machine learning model. A first output is generated from the machine learning model by executing the machine learning model using the constraint vector as a first input to the machine learning model. The constraint vector is converted into a legal action mask. A probability vector is generated by executing a masked softmax operator. The masked softmax operator takes, as a second input, the first output. The masked softmax operator takes, as a third input, the legal action mask. The masked softmax operator generates, as a second output, the probabilities vector. Action outputs are generated by applying a sampling system to the probability vector. The action outputs include a subset of the user actions, and wherein the subset includes only allowed user actions.

METHOD TO IDENTIFY AND MODIFY NWDAF DEPENDENCIES

Resumen de: US20260222304A1

0000 A computer-implemented method performed by a computing device including a NWDAF having a function to identify and modify dependencies between NWDAFs and at least a first NWDAF consumer. The method includes identifying a relationship between at least the first NWDAF consumer from a plurality of NWDAF consumers that respectively subscribe to the NWDAFs that have respective machine learning, ML, models that have an overlap between a first input feature and respectively have different outputs. At least the first NWDAF consumer uses a first output from the NWDAFs to output an action from the first NWDAF consumer. The method further includes determining whether a second input feature to (i) a respective ML model of the NWDAFs or (ii) a new ML model of an unsubscribed NWDAF can make a positive contribution to the prediction of the output of at least the first NWDAF consumer.

CLEANING A MACHINE LEARNING DATASET USING CORESETS

Resumen de: US20260220535A1

A computerized method of cleaning a data set comprises: (a) providing the data set, configured for training of a machine learning model, and comprising data instances. Each instance comprises a label; (b) selecting a number of data instances from the set based on importance-related criteria, indicative of corresponding high importance values associated with the data instances. This generates high importance set(s), giving rise to a set of selected data instances having a high probability of high importance values, as compared to a second probability associated with selecting data instances from the data set. The selected set facilitates determining, for each data instance of, whether it requires an action. This facilitates performing the action, which brings about increased quality data set(s), which facilitate quicker and/or higher-accuracy training of the machine learning model, as compared to a second training of the machine learning model performed utilizing the data set.

BATTERY LIFE ESTIMATION APPARATUS AND METHOD

Resumen de: US20260219323A1

A battery life estimation apparatus includes a charging/discharging unit configured to charge/discharge a battery and a controller configured to calculate a partial accumulative capacity by charging/discharging the battery in a partial voltage period corresponding to a period from a first voltage to a second voltage, and estimate a life corresponding to an entire voltage period of the battery by inputting the first voltage, the second voltage, and the partial accumulative capacity to an estimation model trained based on machine learning.

DETERMINING CONFIDENCE TO START VALUES FOR POWER GENERATION SYSTEMS USING MACHINE LEARNING MODELS

Nº publicación: US20260221775A1 30/07/2026

Solicitante:

CUMMINS POWER GENERATION LTD [GB]
Cummins Power Generation Limited

Resumen de: US20260221775A1

0000 Presented herein are systems and methods for applying machine learning (ML) models to determine start confidence values for power generators. The computing system can identify a first plurality of parameters of a first power generator. The plurality of parameters can identify operations of the first power generator. The computing system can apply the first plurality of parameters to a ML model to determine a first confidence value identifying a first likelihood of the first power generator to start upon initiation. The computing system can provide an output based on the first confidence value for the first power generator.