Absstract of: US2024281966A1
Machine learning systems and methods are disclosed for prediction of wound healing, such as for diabetic foot ulcers or other wounds, and for assessment implementations such as segmentation of images into wound regions and non-wound regions. Systems for assessing or predicting wound healing can include a light detection element configured to collect light of at least a first wavelength reflected from a tissue region including a wound, and one or more processors configured to generate an image based on a signal from the light detection element having pixels depicting the tissue region, determine reflectance intensity values for at least a subset of the pixels, determine one or more quantitative features of the subset of the plurality of pixels based on the reflectance intensity values, and generate a predicted or assessed healing parameter associated with the wound over a predetermined time interval.
Absstract of: WO2024170093A1
There are provided measures for assessment of trustworthiness of a trained machine learning entity. Such measures exemplarily comprise generating at least one artificial intelligence or machine learning trustworthiness sub-factor related metric based on an artificial intelligence or machine learning trustworthiness related behavior of an artificial intelligence or machine learning model, and transmitting an artificial intelligence or machine learning trustworthiness assessment information message comprising said at least one artificial intelligence or machine learning trustworthiness sub-factor related metric.
Absstract of: US2024281713A1
Introduced here is a process for training a machine learning model to predict a characteristic of a textile through analysis of spectral information. The spectral information may include spectra generated by a spectroscopy instrument, for example. The machine learning model can be used for a variety of applications, including automated sorting of textiles for recycling and automated analyzing of textiles for quality control.
Absstract of: US2024281542A1
A machine learning computing system identifies a vulnerability associated with a server. Based on information associated with the server and a knowledge base, the computing system schedules an interval for patching the server in a centralized tracking module. Based on the knowledge base and the vulnerability, the computing system creates, validates, and deploys the patch job. During patch job execution, the computing system monitors the status of the patch job at the server and transmits status updates to a user interface module. After expiration of the interval, the computing system generates an assessment report for the executed patch job. The computing system updates the knowledge base based on the assessment report to improve future decisioning processes. Based on the success or failure of the patch job, the computing system, upon a failure indication, automatically reschedules an interval for patching the server.
Absstract of: US2024281701A1
In some aspects, a computing system may generate uninformative features that may be added to a dataset of real features to use as a baseline for determining the quality of an explanation of model output. The uninformative features may be features that do not correlate with what a model is tasked with predicting (e.g., the uninformative features may be random values), and the real features may be informative and correlate with what the model is tasked with predicting (e.g., variables of a dataset sample). A machine learning model may be trained on a dataset that includes both the real features and the uninformative features. The computing system may generate feature attributions for model output, which may include feature attributions for the uninformative features and the real features in the dataset.
Absstract of: US2024281658A1
Disclosed is a new location threat monitoring solution that leverages deep learning (DL) to process data from data sources on the Internet, including social media and the dark web. Data containing textual information relating to a brand is fed to a DL model having a DL neural network trained to recognize or infer whether a piece of natural language input data from a data source references an address or location of interest to the brand, regardless of whether the piece of natural language input data actually contains the address or location. A DL module can determine, based on an outcome from the neural network, whether the data is to be classified for potential location threats. If so, the data is provided to location threat classifiers for identifying a location threat with respect to the address or location referenced in the data from the data source.
Absstract of: US2024273387A1
A machine learning (ML) system includes a student ML system, a learning coach ML system, and a reference system that generates training data for the student ML system. The learning coach ML system learns to make an enhancement to the student ML system or to its learning process, such as updated hyperparameter or a network structural change, based on training of the student ML system with the training data generated by the reference system. The system may also comprise a learning experimentation system that communicates with the reference system to conduct experiments on the learning of the student learning system. Also, the learning experimentation system can determine a cost function for the learning coach ML system.
Absstract of: US2024273400A1
A hyperparameter tuning system generates, for each hyperparameter, a performance attribution statistic corresponding to an evaluation metric of the machine learning model based on historical experiment statistics for the evaluation metric and the machine learning model. The hyperparameter tuning system allocates a weight to each hyperparameter based on the performance attribution statistic of the hyperparameter. The hyperparameter tuning system updates, in a series of experiments, the hyperparameters based on the weight assigned to each hyperparameter and selects a set of the hyperparameters for the machine learning model from one of the experiments, wherein the set of the hyperparameters results in a recorded value of the evaluation metric that satisfies a tuning condition.
Absstract of: US2024273388A1
An image or a spectrum of a surface may be acquired by a computing device, which may be included in a mobile device in some examples. The computing device may extract a measured spectrum from the image and generate a corrected spectrum of the surface. In some examples, the corrected spectrum may be generated to compensate for ambient light influence. The corrected spectrum may be analyzed to provide a result, such as a diagnosis or a product recommendation. In some examples, the result is based, at least in part, on a comparison of the corrected spectrum to reference spectra. In some examples, the result is based, at least in part, on an inference of a machine learning model.
Absstract of: US2024273372A1
This disclosure relates to artificial intelligence (AI) and machine learning networks for predicting or determining demand metrics across multiple channels. An analytics platform can receive channel events from multiple channels corresponding to geographic areas, and channel features related to demand conditions in the channels can be extracted from the channel events. During a training phase, the channel features can be accumulated into one or more training datasets for training one or more demand prediction models. The one or more demand prediction models can be trained to predict or determine demand metrics for each of the channels. The demand metrics can indicate or predict demand conditions based on the current conditions in the channels and/or based on future, predicted conditions in the channels. Other embodiments are disclosed herein as well.
Absstract of: US2024273389A1
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.
Absstract of: US2024273299A1
Systems and methods associated with generation and/or provision of predictive content are disclosed. One exemplary method includes receiving communications associated with a plurality of customers; determining a message type for each message of the communications; splitting first messages of the first message type into a first set of subcomponent text sections; splitting second messages of the second message type into a second set of subcomponent text sections; analyzing the first set and the second set to generate a plurality of semantic numerical scores for each respective subcomponent text section; determining at least one impactful semantic category for a target audience by selecting at least one semantic category corresponding to at least one semantic numerical score of at least one subcomponent text section of the first set or the second set that is equal to or higher than a first pre-determined threshold value; generating personalized textual content targeting the audience.
Absstract of: US2024273124A1
Some embodiments may perform operations of a process that includes obtaining a natural language text document and use a machine learning model to generate a set of attributes based on a set of machine-learning-model-generated classifications in the document. The process may include performing hierarchical data extraction operations to populate the attributes, where different machine learning models may be used in sequence. The process may include using a pre-trained Bidirectional Encoder Representations from Transformers (BERT) model augmented with a pooling operation to determine a BERT output via a multi-channel transformer model to generate vectors on a per-sentence level or other per-text-section level. The process may include using a finer-grain model to extract quantitative or categorical values of interest, where the context of the per-sentence level may be retained for the finer-grain model.
Absstract of: WO2024168127A1
A method including classifying a cluster training dataset by, for each datapoint of the cluster training dataset: obtaining a student prediction from each of a plurality of respective student machine-learning models for each a plurality of nodes of a cluster; performing a voting of the student predictions from the plurality of respective student machine-learning models of at least a portion of the plurality of nodes of the cluster to determine a respective classification for the datapoint; and labeling the datapoint of the cluster training dataset with the respective classification. The method also can include training a cluster machine-learning model of the cluster using the cluster training dataset, as classified. Other embodiments are described.
Absstract of: US2024274256A1
A system and method to determine a recommendation to change a treatment regimen of a respiratory ailment. The treatment regimen includes multiple steps. Use data of a respiration medicament device to deliver controller or rescue respiration medicament to a patient is collected via a communication interface. The use data is transmitted to a storage device. The use data in the storage device is made accessible to a data analysis module. A patient value associated with the treatment regimen is determined based on the collected use data and patient context data. A comparison of the patient value is made to a threshold level. A recommendation to change the step of the treatment is made based on the comparison to the threshold level. A notification of recommendation of the change of the step of the treatment regimen is provided.
Absstract of: US2024274285A1
A wearable electronic device, a system and methods of monitoring with a wearable electronic device. The device includes a hybrid wireless communication module with wireless communication sub-modules to selectively acquire location data from both indoor and outdoor sources, as well as a wireless communication sub-module to selectively transmit a cellular-based LPWAN signal to provide location information based on the acquired data. The device may also include sensors to collect one or more of environmental, activity and physiological data. The device may transmit some or all of its acquired data to the system to provide a predictive model to correlate changes in the acquired data to corresponding health, safety or related changes to a wearer of the device. In one form, the predictive health care protocol uses a machine learning model at least some of which may be performed on the device.
Absstract of: US2024274105A1
According to an embodiment, there is provided a system and method for automatic AI-based song construction based on ideas of a user. It provides and benefits from a combination of expert knowledge resident in an expert engine which contains rules for a musically correct song generation and machine learning in an AI-based audio loop selection engine for the selection of fitting audio loops from a database of audio loops. Additionally, in some embodiments there is provided a method of energy-based song construction where the tracks of a multi-track work are balanced depending on the desired output volume level of the final project.
Absstract of: US2024273416A1
Methods and apparatus for addressing intents using machine learning (ML) are provided. A method of operation for a node implementing ML, wherein the node instructs actions in an environment in accordance with a policy generated by a ML agent, and wherein the ML agent models the environment, includes obtaining an intent, wherein the intent specifies one or more criteria to be satisfied by the environment. The method further includes determining an intent cluster from among a plurality of intent clusters to which the intent maps, the determination being based on the criteria specified by the intent, and setting initialisation parameters for a ML model to be used to model the intent, based on the determined intent cluster. The method also includes training the ML model using training data specific to the intent, and generating one or more suggested actions to be performed on the environment using the trained ML model.
Absstract of: US2024274106A1
According to an embodiment, there is provided a system and method for automatic AI-based song construction based on ideas of a user. It provides and benefits from a combination of expert knowledge resident in an expert engine which contains rules for a musically correct song generation and machine learning in an AI-based audio loop selection engine for the selection of fitting audio loops from a database of audio loops. Additionally, in some embodiments there is provided a method of energy-based song construction where the tracks of a multi-track work are balanced depending on the desired output volume level of the final project.
Absstract of: US2024273370A1
Systems and methods, for selecting a neural network for a machine learning (ML) problem, are disclosed. A method includes accessing an input matrix, and accessing an ML problem space associated with an ML problem and multiple untrained candidate neural networks for solving the ML problem. The method includes computing, for each untrained candidate neural network, at least one expressivity measure capturing an expressivity of the candidate neural network with respect to the ML problem. The method includes computing, for each untrained candidate neural network, at least one trainability measure capturing a trainability of the candidate neural network with respect to the ML problem. The method includes selecting, based on the at least one expressivity measure and the at least one trainability measure, at least one candidate neural network for solving the ML problem. The method includes providing an output representing the selected at least one candidate neural network.
Absstract of: US2024273412A1
A computer-implemented method of determining whether a task can be completed by machine learning is described. The method comprising the following steps. First of all, test data for the task is obtained. Using the test data, a determination is made for a plurality of machine learning algorithms whether any of the machine learning algorithms is able to perform the task to meet a performance threshold. If none of the machine learning algorithms performs the task to the performance threshold, a set of failure modes are identified, and a determination is made for each failure mode of a likelihood of that failure mode causing failure to meet the performance threshold. From this, an output is provided indicating relative likelihoods of each failure mode of the set causing failure to meet the performance threshold. A computer system suitable for performing the method is also described.
Absstract of: US2024273408A1
A computer implemented method of monitoring pain of an object being estimated based on brainwave data of the object includes a) reading out, by a processor, from a memory a plurality of groups of the brainwave data respectively corresponding to a plurality of levels of stimulations to the object applied by a stimulation system, wherein each group of brainwave data are measured by using an electroencephalograph; b) dividing each group of brainwave data into subgroups each having a first predetermined time frame and obtaining a temporal change of mean values of respective subgroups of the brainwave data, wherein each of the mean values is calculated based on the brainwave data of the first predetermined time frame; and c) evaluating or monitoring a level of pain of the object being estimated from the brainwave data based on the temporal change of the mean values.
Absstract of: US2024273390A1
The present disclosure describes systems and methods for determining correlation codes for tree-based decisioning models. In one embodiment, a method for determining correlation codes in a tree-based decision model includes: assigning each decision node in a tree-based decision model to a correlation code; initializing a risk sum for each correlation code; calculating, for all decision nodes in the tree-based decision model, a difference in risk between child nodes and respective parent nodes; updating the risk sum for each correlation code associated with the decision node used in the decision for the node; determining the feature with the highest risk sum; and determining the correlation code associated with the determined decision node.
Absstract of: AU2022360356A1
In some aspects, techniques for creating representative and informative training datasets for the training of machine-learning models are provided. For example, a risk assessment system can receive a risk assessment query for a target entity. The risk assessment system can compute an output risk indicator for the target entity by applying a machine learning model to values of informative attributes associated with the target entity. The machine learning model may be trained using training samples selected from a representative and informative (RAI) dataset. The RAI dataset can be created by determining the informative attributes based on attributes used by a set of models and further extracting representative data records from an initial training dataset based on the determined informative attributes. The risk assessment system can transmit a responsive message including the output risk indicator for use in controlling access of the target entity to an interactive computing environment.
Nº publicación: EP4414902A2 14/08/2024
Applicant:
HARTFORD STEAM BOILER INSPECTION AND INSURANCE COMPANY [US]
Hartford Steam Boiler Inspection and Insurance Company
Absstract of: EP4414902A2
Systems and methods include processors for receiving training data for a user activity; receiving bias criteria; determining a set of model parameters for a machine learning model including: (1) applying the machine learning model to the training data; (2) generating model prediction errors; (3) generating a data selection vector to identify non-outlier target variables based on the model prediction errors; (4) utilizing the data selection vector to generate a non-outlier data set; (5) determining updated model parameters based on the non-outlier data set; and (6) repeating steps (1)-(5) until a censoring performance termination criterion is satisfied; training classifier model parameters for an outlier classifier machine learning model; applying the outlier classifier machine learning model to activity-related data to determine non-outlier activity-related data; and applying the machine learning model to the non-outlier activity-related data to predict future activity-related attributes for the user activity.