A method includes obtaining a set of image segment identigens for each image segment of an image to produce sets of image segment identigens. The sets of image segment identigens are possible interpretations of an image segment. The method further includes generating a set of relationships between image segments. The relationships provide a list of one or more ways in which the image segments are related. The method further includes processing different permutations of the sets of image segment identigens in accordance with the set of relationships to generate an entigen group. The entigen group represents a most likely interpretation of the image.

A device may receive data that includes invoice data related to historical invoices from an organization, contact data related to historical contacts between the organization and various entities, and dispute data related to historical disputes between the organization and the various entities. The device may determine a profile for the data. The device may determine a set of supervised learning models for the historical invoices based on one or more of the historical contacts, the historical disputes, the historical invoices, or historical patterns related to the historical invoices. The device may determine, using the profile, a set of unsupervised learning models for the historical invoices independent of the one or more of the historical contacts, the historical disputes, or the historical patterns. The device may determine, utilizing a super model, a prediction for the invoice after the super model is trained. The device may perform one or more actions.

Privacy-preserving federated learning apparatuses, systems, computer program products, and methods are provided that generate an updated global model based on a set of client models while maintaining privacy regarding the data values embodying each client model and the updated global model. In this regard, masked client models are utilized, which cryptographically obfuscate data values embodying the client model while still enabling combination, or “aggregation,” of the masked client models to generate a masked updated global model. The masked updated global model similarly includes obfuscated data values embodying the updated global model, but may be unmasked to reveal the true values of the updated global model for use. Some embodiments utilize specific steps for communication between environments, systems, devices, and/or the like, to ensure the masked models can only be unmasked by intended entities.

A controlling method of an electronic device are provided. The controlling method includes obtaining, a first feature value corresponding to the input data, mapping information included in the first feature value to a symbolic value corresponding to each of a logically certain information and uncertain information, tweaking a gradient corresponding to the uncertain information, obtaining a second feature value in which the first feature value and the symbolic value are merged, obtaining temporary answer information including the certain information and the uncertain information, obtaining score information related to a level the temporary answer information matches a pre-defined logic rule, training the neural network model based on the tweaked gradient when the score information is less than a pre-set threshold value, and obtaining an output data on the input data based on the obtained second feature value when the score information is greater than or equal to the pre-set threshold value.

Methods and systems for learning policies using sparse and underspecified rewards. One of the methods includes training the policy jointly with an auxiliary reward function having a plurality of auxiliary reward parameters, the auxiliary reward function being configured to map, in accordance with the auxiliary reward parameters, trajectory features of at least a trajectory to an auxiliary reward value that indicates how well the trajectory performed a task in response to a context input.

A recognition processing section performs subject recognition in a processing area of an image obtained by an imaging section. The recognition processing section determines an image characteristic of the processing area on the basis of a characteristic map indicating an image characteristic of the image obtained by the imaging section and uses a recognizer corresponding to the image characteristic of the processing area. The characteristic map includes a map based on an optical characteristic of an imaging lens used in the imaging section and is stored in a characteristic information storage section. An imaging lens has a winder angle of view in all directions or in a predetermined direction than a standard lens, and the optical characteristic thereof differs depending on a position on the lens. The recognition processing section performs the subject recognition using a recognizer corresponding to resolution or skewness of the processing area, for example.

An exemplary system, method, and computer-accessible medium can include, for example, establishing a unique rule-identifier in one-to-one correspondence with at least one set of unknown time-variable rules against which data is to be made compliant, obtaining at least one set of data marked compliant against the one or more set of rules, obtaining meta-data from the compliant data, obtaining at least one set of data marked non-compliant against the set of unknown time-variable rules, extracting meta-data from the non-compliant data, joining the set of compliant and non-compliant metadata to generate a set of estimated rules corresponding to the rule-identifier based at least one of (i) the meta-data of the joined set and (ii) machine learning algorithms.

The present invention relates generally to identifying relationships between items. Certain embodiments of the present invention are configurable to identify the probability that a certain event will occur by identifying relationships between items. Certain embodiments of the present invention provide an improved supervised machine learning system.

Presented are concepts for matching a subject to one or more resources or workflow steps. Once such concept comprises obtaining data associated with a subject, the data comprising, for each of a plurality of parameters, a parameter value relating to the subject. A plurality of data groups for characterising the subject is then generated and a classification process is applied to each data group so as to generate a classification result for each data group. The subject is then matched to one or more resources or workflow steps based on the classification results.

Examples disclosed herein relate to determining a response based on hierarchical models. In one implementation, a processor applies a first model to an image of an environment to select a second model. The processor applies the selected second model to the image and creates an environmental description representation based on the output of the second model. The processor determines a response based on the environmental description information.