Alerta

MACHINE LEARNING BASED SURFACE NETWORK MODELING

Resumen de: US20260203483A1

0000 The present disclosure describes techniques including sampling a multidimensional physical space to determine representative combinations of pipeline segment parameters, and executing a flow simulator to estimate pressure drop values or pressure gradient values for pipeline segments having the representative combinations of pipeline segment parameters. The techniques also include training a ML model using the pressure drop values or the pressure gradient values estimated by the flow simulator. The ML model may output a predicted pressure drop value or pressure gradient value for a pipeline segment based on input values representing pipeline segment parameters of the pipeline segment. The techniques can also include upscaling a network model before executing a ML-based network solver to estimate a pressure drop value, a flow rate value, and node pressure values for the network model using the trained ML model.

MONOCULAR DEPTH AND OPTICAL FLOW ESTIMATION USING DIFFUSION MODELS

Resumen de: US20260203923A1

Improved methods are provided for generating, via a noise-diffusion iterative process, depth maps or optical flow maps from input images. Also provided are improved methods for training the machine learning model(s) employed in the iterative process and for augmenting die set of training data used to train such models. By translating the depth or optical flow map prediction process into the noise diffusion context, improved performance with respect to compute cost, training data, requirements, model size, and output quality are obtained. Additionally, the noise diffusion context allows models trained as described herein to generate maps de novo from target color images and/or to begin from initial ‘guess’ maps (e.g., noisy maps, maps containing holes) when generating improved output maps, natively incorporating the imperfect prior information represented by such initial maps.

USER INTERFACE FOR MATERIALS AND MATERIAL PROPERTIES PREDICTION USING A MACHINE LEARNING MODEL

Resumen de: US20260204366A1

0000 Machine learning can be used to predict formulations for an output formulation. The machine learning can be implemented by a machine learning model, which employs a forward model and an inverse model. A user interface can be used to gather raw materials selections and output formulation property selections. The selections can be used to generate formulations that comply with selections using the ML model.

DEVICES, METHODS, APPARATUSES AND COMPUTER READABLE MEDIUM FOR COMMUNICATIONS

Resumen de: US20260205212A1

0000 Embodiments of the present disclosure disclose devices, methods and apparatuses for communications. In the embodiments, a network device receives at least one of reference signal received power (RSRP) and reference signal received quality (RSRQ) associated with a serving cell of a terminal device from the terminal device. Then, the network device determines, based on the at least one of the RSRP and the RSRQ associated with the serving cell, a signal quality level associated the neighboring cell of the serving cell using a machine learning (ML) or artificial intelligence (AI) model. In this way, the throughput of the communication system can be improved.

APPARATUS AND METHODS FOR CHARACTERIZATION OF MULTIPLE ELECTRIC MOTORS

Resumen de: US20260203472A1

The techniques described herein relate to systems and methods for characterization of multiple electric motors. An example method for processing multiple electric motor designs into outputs of respective performance evaluations across different operating conditions using machine learning includes mapping input geometric parameters to at least one of a plurality of electric motor designs, and inputting the plurality of electric motor designs to at least one machine learning model and outputting, from the at least one machine learning model, performance evaluations for the plurality of electric motor designs under a variety of operating conditions, the at least one machine learning model trained to generate the performance evaluations in accordance with control waveforms for the plurality of electric motor designs.

Optimization Method for Distributed Execution of Deep Learning Task and Distributed System

Resumen de: US20260203135A1

Disclosed are an optimization method for distributed execution of a deep learning task and a distributed system. The method includes that: a computation graph is generated based on a deep learning task and hardware resources are allocated for the distributed execution of the deep learning task; the allocated hardware resources are grouped to obtain at least one grouping scheme; for each grouping scheme, tensor information related to multiple operators contained in the computation graph is split based on the value of at least one factor under this grouping scheme to obtain multiple candidate splitting solutions; and an optimal efficiency solution for executing the deep learning task of the hardware resources is selected by using a cost model. Through operator splitting based on device grouping combined with optimization solving based on the cost model, automatic optimization of distributed execution for various deep learning tasks is realized.

SYSTEM AND METHOD FOR HIGH-THROUGHPUT EXECUTION OF DEEP-LEARNING WORKLOADS

Resumen de: WO2026148416A1

A system and method of executing a deep-learning workload are provided. The method includes selecting at least one parallelization strategy for a first processing stage of the workload, selecting at least one different parallelization strategy for a second processing stage of the workload, and switching between the different strategies used for the first and second processing stages by reconfiguring a partitioning of the deep-learning workload across computing resources.

CONTROL PARAMETER BASED SEARCH SPACE FOR VEHICLE MOTION PLANNING

Resumen de: US20260200496A1

0000 A method for generating a trajectory for a vehicle based on an abstract space of control parameters associated with the vehicle may include applying a machine learning model to determine an abstract space representation of the trajectory, which includes a sequence of control parameters associated with the vehicle. For instance, application of the machine learning model may include performing a search of the abstract space parameterized by control parameters that include derivatives of the position of the vehicle. Examples of control parameters include velocity, acceleration, jerk, and snap. A physical space representation of the trajectory may be determined by at least mapping the sequence of control parameters to a sequence of positions for the vehicle. A motion of the vehicle may be controlled based at least on the physical space representation of the trajectory. Related systems and computer program products are also provided.

METHOD FOR REPORTING INDICATION OF INVALID INFERENCE/PREDICTION REPORT

Resumen de: WO2026151371A1

A method, system and apparatus are disclosed A method implemented in a user equipment configured to communicate with a network node includes: generating (S204), based on a machine learning, ML, model/functionality, a report for one or both of an inference report and a prediction report, the report comprising an indication that the one or both of the inference report and the prediction report is one or more of invalid, out of range, and inaccurate; and transmitting (S206) the report to the network node.

CONTINUOUS ACCESS CONTROL GROUP DETERMINATION

Resumen de: WO2026151584A1

A data management server may receive metadata related to data access associated with a domain containing a plurality of named entities. The server determines domain-defined groups, each associated with a domain-defined access rule, and extracts features from the metadata, including access activity data of the named entities accessing domain resources. These features are input into a machine learning model to identify an automatically-generated access-control group comprising one or more named entities. The server determines an access control rule for the generated group based on access activity data and compares it to a domain-defined access rule for a domain-defined grouping containing one of the named entities. Based on this comparison, the server generates a recommendation to modify access privileges for one or more named entities in the group, facilitating enhanced access control aligned with domain-specific policies and security objectives.

MODEL CASCADE FOR EFFICIENT RULE EXTRACTION

Resumen de: US20260203056A1

0000 A method including determining whether to use a functional processing model or a machine-learning processing model and executing the functional processing model or the machine-learning processing model using as input the first input document to transform data in a first input document into the target data format, executing an extraction model using as input the data in the first input document in the target data format to generate first structured data, executing a compilation model using as input a plurality of structured data generated based on data in a set of input documents including the first input document to generate an aggregate structured data, and executing a synthesis model using as input the aggregate structured data and the plurality of structured data to generate an output data structure.

MACHINE-LEARNING TECHNIQUES FOR RISK ASSESSMENT BASED ON CLUSTERING

Resumen de: US20260203611A1

Systems and methods for predicting future risk for a target entity are provided. A risk assessment system receives historical risk assessment data of the target entity and identifies a target cluster that matches the historical risk assessment data. The target cluster is identified from a group of clusters determined using high dimensional clustering based on risk assessment data of a set of entities. The risk assessment system identifies a set of nearest neighbors of the target cluster and determines a prediction of future risk for the target entity based on the target cluster and the set of nearest neighbors. The risk assessment system transmits a responsive message, which can include the prediction of future risk, to a remote computing device for use in controlling access of the target entity to one or more interactive computing environments.

ARTIFICIAL-INTELLIGENCE-ENABLED AUTHENTICATION BASED ON USER METADATA

Resumen de: US20260203388A1

A system includes memory hardware configured to store instructions and one or more electronic processors configured to execute the instructions. The instructions include logging historical behavioral biometric metadata from one or more computing platforms, generating a user profile based on the logged historical behavioral biometric metadata, receiving an authentication request from a client computing platform, providing the first behavioral biometric metadata and the user profile to a trained machine learning model to generate a biometric match, generating a positive control signal in response to a positive biometric match, sending the positive control signal to the client computing platform, updating the historical behavioral biometric metadata with the first behavioral biometric metadata, and retraining the trained machine learning model using the updated historical behavioral biometric metadata. The authentication request includes first behavioral biometric metadata.

VARIABLE-FOCUS DYNAMIC VISION FOR ROBOTIC SYSTEM

Resumen de: US20260200090A1

0000 A dynamic vision system for a robotic system includes an optical assembly including a lens containing a liquid. The lens is deformable to generate variable focus for the lens. The optical assembly is configured to capture optical data. A robotic system is configured to simulate human or animal species capabilities having a control system configured to adjust one or more optical parameters. The one or more optical parameters modify the variable focus of the lens while the optical assembly captures current optical data relating to the robotic system. A processing system is configured to train a machine learning model to recognize an object relating to the robotic system from training data generated from the optical data captured by the optical assembly. The optical data includes the current optical data relating to the robotic system.

METHOD FOR EXTRACTING ENTITIES AND RELATIONSHIPS FROM A CORPUS TO POPULATE A KNOWLEDGE GRAPH

Resumen de: EP4776154A1

The present invention discloses a method for extracting entities and relations from technical documents in a mostly automated way that achieves a more complete and accurate result in a reduced time frame. Several deep learning models are trained using a corpus of the domain of interest annotated by experts and linguists. A graph vector model is also trained. The manual annotations and revisions are the minimum necessary to obtain automated models capable of automatically extracting the entities and relations from a corpus. Once trained, the models can be used in any corpus within the same domain of knowledge.

DETECTION OF ELECTRIC VEHICLE CHARGING AT PREMISE

Resumen de: EP4775440A2

Systems, methods, and other embodiments associated with detecting an electric vehicle charging event. In one embodiment, from electricity consumption data from a known set of electric vehicle owners, the method encodes usage values from time intervals with a symbol from a series of symbols representing a level of electricity consumption during the time interval. The encoding generates an encoded consumption pattern of symbols for each electrical vehicle owner. An EV charge motif is identified that represents an EV charging event. One or more machine learning classifiers is trained to identify the EV charge motifs from the known set of electric vehicle owners and to distinguish from non-charge motifs to identify EV charges from unknown data sets.

DIFFERENTIALLY PRIVATE FEDERATED EXTREME GRADIENT BOOSTING

Resumen de: US20260195639A1

Training a differential privacy-aware (DP-aware) machine learning model includes transmitting epsilon hyperparameters to federated learning (FL) nodes. A differential privacy-aware (DP-aware) machine learning model is generated based on noise-infused surrogate histograms received from the FL nodes, each noise-infused surrogate histogram based on an epsilon hyperparameter and representing a node-specific dataset. The DP-aware machine learning model is transmitted to the FL nodes. A DP-aware aggregate histogram is generated by merging DP-aware gradients and DP-aware Hessians determined by the FL nodes based on each FL node generating predictions by applying the DP-aware machine learning model to a node-specific dataset therein. A decision tree of the DP-aware machine learning model is expanded by dividing data in one or more decision tree nodes. The machine learning model is iteratively trained by successively merging further DP-aware gradients and DP-aware Hessians generated by FL nodes based on updated versions of the DP-aware machine learning model.

WEATHER MODELLING

Resumen de: US20260195659A1

A computer-implemented method of training a first machine learning model to determine a local weather modifier, the method comprising: obtaining a first training data set comprising a plurality of training samples, each training sample comprising an observed local weather modifier in a geographical region and one or more geographic features of the geographical region, wherein each geographic feature has a feature location and a feature property; training the first machine learning model to predict a local weather modifier for another geographical region based on one or more geographic features of the other geographical region.

SYSTEM AND METHOD FOR PREDICTING SHOE TRACTION USING COMPUTER VISION

Resumen de: US20260191300A1

0000 In one embodiment, a method includes accessing an image depicting a portion of a shoe and a background of the shoe, segmenting the portion of the shoe from the background by machine-learning models, extracting features configured for traction prediction by the machine-learning models, and determining a traction performance associated with the shoe based on the extracted feature by the machine-learning models.

AUTOMATED DATA CLASSIFICATION

Resumen de: US20260195416A1

Automatically classifying data fields given a set of their sample values (e.g., schema mapping) is disclosed. These models aim to infer attribute and entity structure defined by a platform and/or automated classification system. These models may rely upon both public machine learning models as well as proprietary model weights owned by the platform and/or automated classification system (and/or associated organizations or enterprises).

SYSTEMS AND METHODS FOR GENERATING CUSTOMIZED TRAINING

Resumen de: US20260195418A1

A system may be configured to perform a method for generating customized training. The system may receive first user interaction data associated with a user. The system may determine, using a machine learning model (MLM), whether the first user interaction data exceeds a predetermined threshold. Based on such determination, the system may assign a training module to the user. The system may access a user profile associated with the user, the user profile comprising a plurality of training modules. The system may generate a training plan based on the training module and the plurality of training modules. The system may receive second user interaction data associated with the user, and may determine an efficacy level of the training plan based on the second user interaction data. The system may dynamically update the training plan based on the efficacy level, and may dynamically display the training plan in the user profile.

Federated Learning applications for secure and private Machine Learning in Oil and Gas Industry

Resumen de: US20260195608A1

0000 Techniques for training a global model for use at a host of oilfield application sites based on raw data obtained from the application sites without direct exposure of the data to the global model. The techniques include developing and distributing a global model with a predetermined set of parameter weights. The model is then locally employed at each application site by a local computer which maintains the integrity of the acquired data during performance of the oilfield application. The data is used to update the parameter weights based on real-time circumstances. Thus, the parameter weights may be transmitted to the centralized computer for updating of the global model. Further, the updated global model may continue to direct other applications and the process continued in a beneficial feedback loop manner.

ANOMALY DETECTION USING GRAPH NEURAL NETWORKS

Resumen de: US20260197331A1

0000 Example implementations relate to detecting a terminated entity in a network environment. A network activity dataset including data representative of network activity within a network environment and a plurality of data records is received. Each data record in the plurality of data records includes a set of attributes. A graph that links systems having a first role in the data representative of network activity and a subset of the plurality of data records is generated. Feature information from the set of attributes for one or more data records in the subset of the plurality of data records in the graph is aggregated. A machine learning model is trained based on the aggregated feature information derived from the graph. Using the trained model, a determination representing a likelihood that a respective system having the first role in the data representative of network activity is linked to the terminated entity is generated.

DEEP LEARNING-BASED PATH PLANNING METHOD CONSIDERING HULL STABILITY OF UNMANNED SURFACE VEHICLE IN RESPONSE TO MARITIME ENVIRONMENT AND DEVICE THEREFOR

Resumen de: KR20260108793A

본 발명의 다양한 실시예에 따르면, 무인수상정의 선체 안정성을 고려한 딥러닝 기반 경로계획 장치는 상기 무인수상정의 내부에 위치한 센서 또는 외부로부터 상기 무인수상정의 상태 정보 또는 해상 환경 정보를 입력받고, 무인수상정의 상태 정보 또는 해상 환경 정보를 입력받아 무인수상정 롤(Roll) 정보를 출력하도록 미리 학습된 무인수상정 전복 위험도 예측 모델에 상기 입력받은 무인수상정의 상태 정보 또는 해상 환경 정보를 입력하고, 상기 무인수상정 전복 위험도 예측 모델로부터 출력된 상기 무인수상정 롤 정보를 미리 정의된 휴리스틱(heuristic) 함수에 적용하고, 상기 휴리스틱 함수 값을 기반으로 상기 무인수상정의 이동 경로를 결정할 수 있다.

SYSTEMS AND METHODS TO GENERATE DATA MESSAGES INDICATING A PROBABILITY OF EXECUTION FOR DATA TRANSACTION OBJECTS USING MACHINE LEARNING

Nº publicación: US20260195660A1 09/07/2026

Solicitante:

NASDAQ INC [US]
Nasdaq, Inc.

Resumen de: US20260195660A1

A computer system includes a transceiver that receives over a data communications network different types of input data and multiple data transaction objects from multiple source nodes. A pre-processor processes the different types of input data and the data transaction objects to generate an input data structure. Based on the input data structure, one or more predictive machine learning models is trained and used to predict a probability of execution of each of the data transaction objects at a future execution time. Output data messages are then generated for transmission by the transceiver over the data communications network indicating the probability of execution for at least one of the data transaction objects at the future execution time.