Alerta

ADAPTIVE LENGTH SPECULATIVE DECODING IN AUTOREGRESSIVE GENERATIVE ARTIFICIAL INTELLIGENCE MODELS

Resumen de: US2025245530A1

Certain aspects of the present disclosure provide techniques and apparatus for generating a response to a query input in a generative artificial intelligence model using variable draft length. An example method generally includes determining (e.g., measuring or accessing) one or more operational properties of a device on which inferencing operations using a machine learning model are performed. A first draft set of tokens is generated using the machine learning model. A number of tokens included in the first draft set of tokens is based on the one or more operational properties of the device and a defined scheduling function for the machine learning model. The first draft set of tokens are output for verification.

Using Optical Remote Sensors And Machine Learning Models To Predict Agronomic Field Property Data

Resumen de: US2025245532A1

In some embodiments, a computer-implemented method for predicting agronomic field property data for one or more agronomic fields using a trained machine learning model is disclosed. The method comprises receiving, at an agricultural intelligence computer system, agronomic training data; training a machine learning model, at the agricultural intelligence computer system, using the agronomic training data; in response to receiving a request from a client computing device for agronomic field property data for one or more agronomic fields, automatically predicting the agronomic field property data for the one or more agronomic fields using the machine learning model configured to predict agronomic field property data; based on the agronomic field property data, automatically generating a first graphical representation; and causing to display the first graphical representation on the client computing device.

On-Device Machine Learning Platform

Resumen de: US2025245533A1

The present disclosure provides systems and methods for on-device machine learning. In particular, the present disclosure is directed to an on-device machine learning platform and associated techniques that enable on-device prediction, training, example collection, and/or other machine learning tasks or functionality. The on-device machine learning platform can include a context provider that securely injects context features into collected training examples and/or client-provided input data used to generate predictions/inferences. Thus, the on-device machine learning platform can enable centralized training example collection, model training, and usage of machine-learned models as a service to applications or other clients.

SYSTEMS AND METHODS OF OPTIMIZING RESOURCE ALLOCATION USING MACHINE LEARNING AND PREDICTIVE CONTROL

Resumen de: US2025245050A1

A computer system includes a transceiver that receives over a data communications network different types of input data from multiple source nodes and a processing system that defines for each of multiple data categories, a set of groups of data objects for the data category based on the different types of input data. Predictive machine learning model(s) predict a selection score for each group of data objects in the set of groups of data objects for the data category for a predetermined time period. Control machine learning model(s) determine how many data objects are permitted for each group of data objects based on the selection score. Decision-making machine learning model(s) prioritize the permitted data objects based on one or more predetermined priority criteria. Subsequent activities of the computer system are monitored to calculate performance metrics for each group of data objects and for data objects actually selected during the predetermined time period. Predictive machine learning model(s) and decision-making machine learning model(s) are adjusted based on the performance metrics to improve respective performance(s).

Load Balancing For Mixture Of Experts Machine Learning

Resumen de: US2025245553A1

Aspects of the disclosure are directed to improving load balancing for serving mixture of experts (MoE) machine learning models. Load balancing is improved by providing memory dies increased access to computing dies through a 2.5D configuration and/or an optical configuration. Load balancing is further improved through a synchronization mechanism that determines an optical split of batches of data across the computing die based on a received MoE request to process the batches of data. The 2.5D configuration and/or optical configuration as well as the synchronization mechanism can improve usage of the computing die and reduce the amount of memory dies required to serve the MoE models, resulting in less consumption of power and lower latencies and complexity in alignment associated with remotely accessing memory.

TRAINING A MACHINE-LEARNING MODEL FOR CONSTRAINT-COMPLIANCE PREDICTION USING AN ACTION-BASED LOSS FUNCTION

Resumen de: WO2024081965A1

An online system trains a constraint prediction machine-learning model using an action-based loss function. The action-based loss function computes a weighted sum of (1) an accuracy of the constraint prediction model in predicting whether a user's interaction complies with a set of constraints and (2) an action score representing a number of actions taken by users to determine whether the user's action complies with the set of constraints. The online system may apply this trained constraint prediction model to future interaction data received from third-party systems to predict whether user interactions with those third-party systems comply with the set of constraints.

FORWARD-FORWARD TRAINING FOR MACHINE LEARNING

Resumen de: WO2024112887A1

Example implementations provide a computer-implemented method for training a machine-learned model, the method comprising: processing, using a layer of the machine-learned model, positive input data in a first forward pass; updating one or more weights of the layer to adjust, in a first direction, a goodness metric of the layer for the first forward pass; processing, using the layer, negative input data in a second forward pass; and updating the one or more weights to adjust, in a second direction, the goodness metric of the layer for the second forward pass.

MODELLING CAUSATION IN MACHINE LEARNING

Resumen de: US2024104338A1

A method comprising: sampling a temporal causal graph from a temporal graph distribution specifying probabilities of directed causal edges between different variables of a feature vector at a present time step, and from one variable at a preceding time step to another variables at the present time step. Based on this there are identified: a present parent which is a cause of the selected variable in the present time step, and a preceding parent which is a cause of the selected variable from the preceding time step. The method then comprises: inputting a value of each identified present and preceding parent into a respective encoder, resulting in a respective embedding of each of the present and preceding parents; combining the embeddings of the present and preceding parents, resulting in a combined embedding; inputting the combined embedding into a decoder, resulting in a reconstructed value of the selected variable.

MODELLING CAUSATION IN MACHINE LEARNING

Resumen de: US2024104370A1

A method comprising: sampling a first causal graph from a first graph distribution modelling causation between variables in a feature vector, and sampling a second causal graph from a second graph distribution modelling presence of possible confounders, a confounder being an unobserved cause of both of two variables. The method further comprises: identifying a parent variable which is a cause of a selected variable according to the first causal graph, and which together with the selected variable forms a confounded pair having a respective confounder being a cause of both according to the second causal graph. A machine learning model encodes the parent to give a first embedding, and encodes information on the confounded pair give a second embedding. The embeddings are combined and then decoded to give a reconstructed value. This mechanism may be used in training the model or in treatment effect estimation.

LOAD BALANCING FOR MIXTURE OF EXPERTS MACHINE LEARNING

Resumen de: EP4592895A1

Aspects of the disclosure are directed to improving load balancing for serving mixture of experts (MoE) machine learning models. Load balancing is improved by providing memory dies increased access to computing dies through a 2.5D configuration and/or an optical configuration. Load balancing is further improved through a synchronization mechanism that determines an optical split of batches of data across the computing die based on a received MoE request to process the batches of data. The 2.5D configuration and/or optical configuration as well as the synchronization mechanism can improve usage of the computing die and reduce the amount of memory dies required to serve the MoE models, resulting in less consumption of power and lower latencies and complexity in alignment associated with remotely accessing memory.

SYSTEMS AND METHODS FOR PREDICTING HYDRAULIC FRACTURING DESIGN PARMATERS BASED ON INJECTION TEST DATA AND MACHINE LEARNING

Resumen de: US2025237131A1

Systems and methods presented herein include systems and methods for receiving data relating to an injection/falloff test performed in a well in fluid communication with a subterranean reservoir; determining operational parameters of a hydraulic fracturing operation using at least a portion of the data; applying the operational parameters to a pre-trained machine learning predictive model to determine an optimal set of control parameters; and issuing one or more commands relating to the control parameters to optimize the hydraulic fracturing operation on the subterranean reservoir.

SYSTEMS AND METHODS FOR PROVIDING AUTOMATED NATURAL LANGUAGE DIALOGUE WITH CUSTOMERS

Resumen de: US2025239253A1

A system includes one or more memory devices storing instructions, and one or more processors configured to execute the instructions to perform steps of providing automated natural dialogue with a customer. The system may generate one or more events and commands temporarily stored in queues to be processed by one or more of a dialogue management device, an API server, and an NLP device. The dialogue management device may create adaptive responses to customer communications using a customer context, a rules-based platform, and a trained machine learning model.

SYSTEM, METHOD, AND COMPUTER PROGRAM PRODUCT FOR MANAGING AUTOMATED HEALTHCARE DATA APPLICATIONS USING ARTIFICIAL INTELLIGENCE

Resumen de: US2025239366A1

Provided is a system for managing automated healthcare data applications using artificial intelligence (AI) that includes at least one processor programmed or configured to receive healthcare data from a data source, determine a classification of the healthcare data using a machine learning model, wherein the machine learning model is configured to provide a predicted classification of an automated healthcare data analysis application of a plurality of automated healthcare data analysis applications based on an input, and provide the healthcare data to an automated healthcare data analysis application based on the classification of the healthcare data. Methods and computer program products are also disclosed.

Statistical Network Application Security Policy Generation

Resumen de: US2025240329A1

A method for automatically generating network communication policies employs unsupervised machine learning on unlabeled network data representing communications between applications on multiple computer systems. This approach uniquely derives policy rules without predefined labels or user-defined communication categories, ensuring automated rules complement existing user-generated policies by excluding them during training. The method validates network interactions by enforcing rules that leverage application fingerprints and identified feature clusters to distinguish permitted from prohibited communications. Additional techniques include dynamically adapting policies, utilizing decision trees, frequent itemset discovery, and evolutionary algorithms. Suspicious applications are flagged, and malicious data is excluded from training. The system uses aggregated flows, MapReduce processing, and simulated annealing optimization, providing human-readable, periodically retrained rules for balanced network security management.

MANAGING A MODEL TRAINED USING A MACHINE LEARNING PROCESS

Resumen de: US2025238720A1

A computer implemented method of managing a first model that was trained using a first machine learning process and is deployed and used to label medical data. The method comprises determining (202) a performance measure for the first model, and if the performance measure is below a threshold performance level, triggering (204) an upgrade process wherein the upgrade process comprises performing further training on the first model to produce an updated first model, wherein the further training is performed using an active learning process wherein training data for the further training is selected from a pool of unlabeled data samples, according to the active learning process, and sent to a labeler to obtain ground truth labels for use in the further training.

AUTOMATED METHODS AND SYSTEMS THAT PROVIDE RESOURCE RECOMMENDATIONS FOR VIRTUAL MACHINES

Resumen de: US2025238280A1

The current document is directed to methods and systems that generate recommendations for resource specifications used in virtual-machine-hosting requests. When distributed applications are submitted to distributed-computer-system-based hosting platforms for hosting, the hosting requestor generally specifies the computational resources that will need to be provisioned for each virtual machine included in a set of virtual machines that correspond to the distributed application, such as the processor bandwidth, memory size, local and remote networking bandwidths, and data-storage capacity needed for supporting execution of each virtual machine. In many cases, the hosting platform reserves the specified computational resources and accordingly charges for them. However, in many cases, the specified computational resources significantly exceed the computational resources actually needed for hosting the distributed application. The currently disclosed methods and systems employ machine learning to provide accurate estimates of the computational resources for the VMs of a distributed application.

MODULAR SYSTEMS AND METHODS FOR SELECTIVELY ENABLING CLOUD-BASED ASSISTIVE TECHNOLOGIES

Resumen de: US2025238596A1

Systems and methods are disclosed for manually and programmatically remediating websites to thereby facilitate website navigation by people with diverse abilities. For example, an administrator portal is provided for simplified, form-based creation and deployment of remediation code, and a machine learning system is utilized to create and suggest remediations based on past remediation history. Voice command systems and portable document format (PDF) remediation techniques are also provided for improving the accessibility of such websites.

IMPROVED CODE GENERATION USING TREE STRUCTURE FOR TRAINING AND TRIMMING OF GENERATIVE MODEL OUTPUT

Resumen de: EP4589421A1

A method implemented using one or more processors comprises selecting a starting location in an original code snippet, processing the original code snippet to generate a tree representation of the original code snippet, identifying a subtree of the tree representation that contains the starting location in the original code snippet, identifying a ground truth portion of the original code snippet that corresponds to at least a portion of the subtree of the tree representation, and training a machine learning model to generate a predicted code snippet that corresponds to the portion of the subtree. The training includes processing a remainder of the original code snippet outside of the ground truth portion using the machine learning model.

SYSTEM AND METHOD FOR ML-BASED ENGINEERING LIBRARY TRANSLATION AND INTEGRATION, AND FOR SCHEMA & FILE FORMAT MAPPING

Resumen de: EP4589435A1

The present invention relates to a method, comprising: using an artificial intelligence/machine learning, AI/ML, model to map content between an interface of a first entity for interaction with other entities and an interface of a second entity for interaction with other entities, and/or classify content of the interface of the first entity and/or of the interface of the second entity; and obtaining, from the AI/ML model, a first output indicative of a result of the mapping of the content, and/or of a result of the classification of the content.

SYSTEM AND METHOD FOR MACHINE LEARNING BASED GENERATION OF MEDIA STRUCTURED PRODUCTS

Resumen de: US2025232323A1

A method, computer program product, and computer system for identifying, by a computing device, event specific data of a past media program. A size of delivery of a future media program may be predicted based upon, at least in part, the event specific data of the past media program. A financial media product may be generated based upon, at least in part, the predicted size of delivery of the future media program. An actual size of delivery may be identified. A first action may be executed if the actual size of delivery is greater than the predicted size of delivery based upon, at least in part, the financial media product. A second action may be executed if the actual size of delivery is less than the predicted size of delivery based upon, at least in part, the financial media product.

COMPUTER-BASED SYSTEMS FOR DYNAMIC PERSONA-BASED ACCESS TO COMPUTER NETWORK RESOURCES BASED ON MACHINE LEARNING TECHNIQUES AND METHODS OF USE THEREOF

Resumen de: US2025233925A1

At least some embodiments are directed to a system that receives a profile values associated from new user profiles of a computer network or system. A machine learning system determines a set of existing profiles that share at least one common profile value with the new user profile. A second machine learning model determines a set of existing user entitlements associated with the set of existing profiles. The new user profile is processed by a natural language processing engine to determine a set of new user entitlements from the set of existing user entitlements. The system provides the new user with access to electronic resources of the computer network. The system tracks the new user computer network or system activities and updates the new user profile based on the set of new user entitlements and the new user activity on the computer network or system.

System and Method for ML-Based Engineering Library Translation and Integration, and for Schema & File Format Mapping

Resumen de: US2025232119A1

A method includes using an artificial intelligence/machine learning, AI/ML, model to map content between an interface of a first entity for interaction with other entities and an interface of a second entity for interaction with other entities, and/or classify content of the interface of the first entity and/or of the interface of the second entity; and obtaining, from the AI/ML model, a first output indicative of a result of the mapping of the content, and/or of a result of the classification of the content.

SYSTEMS AND METHODS FOR MACHINE LEARNING-BASED DATA MATCHING AND RECONCILIATION OF INFORMATION

Resumen de: US2025231922A1

A method for machine learning-based data matching and reconciliation may include: ingesting a plurality of records from a plurality of data sources; identifying company associated with each of the plurality of records; assigning a unique identifier to each uniquely identified company; matching each of the records to one of the uniquely identified companies using a trained company matching machine learning engine; identifying a primary company record in the matching records and associating other matching records with the primary company record; matching each of the records to a contact using a trained contact matching machine learning engine; identifying a primary contact record in the matching records and associating other matching records with the primary contact record; synchronizing the plurality of records in a graph database using the unique identifier; receiving feedback on the matching companies and/or matching contacts; and updating the trained company matching machine learning engine.

GENERATING AND DEPLOYING PACKAGES FOR MACHINE LEARNING AT EDGE DEVICES

Resumen de: US2025232226A1

A provider network implements a machine learning deployment service for generating and deploying packages to implement machine learning at connected devices. The service may receive from a client an indication of an inference application, a machine learning framework to be used by the inference application, a machine learning model to be used by the inference application, and an edge device to run the inference application. The service may then generate a package based on the inference application, the machine learning framework, the machine learning model, and a hardware platform of the edge device. To generate the package, the service may optimize the model based on the hardware platform of the edge device and/or the machine learning framework. The service may then deploy the package to the edge device. The edge device then installs the inference application and performs actions based on inference data generated by the machine learning model.

METHOD, MEDIUM, AND SYSTEM FOR GENERATING SYNTHETIC DATA

Nº publicación: US2025232353A1 17/07/2025

Solicitante:

CAPITAL ONE SERVICES LLC [US]
Capital One Services, LLC

Resumen de: US2025232353A1

Systems and methods for generating synthetic data are disclosed. A system may include one or more memory devices storing instructions and one or more processors configured to execute the instructions. The instructions may instruct the system to categorize consumer data based on a set of characteristics. The instructions may also instruct the system to receive a first request to generate a first synthetic dataset. The first request may specify a first requirement for at least one of the characteristics. The instructions may further instruct the system to retrieve, from the consumer data, a first subset of the consumer data satisfying the first requirement. The instructions may also instruct the system to provide the first subset of consumer data as input to a data model to generate the first synthetic dataset, and to provide the first synthetic dataset as training data to a machine-learning system.