A computer implemented method of executing a plurality of discrete software modules each including a machine learning algorithm as an executable software component configurable to approximate a function relating a domain data set to a range data set; a data store; and a message handler as an executable software component arranged to receive input data and communicate output data for the module, wherein the message handler is adapted to determine domain parameters for the algorithm based on the input data and to generate the output data based on a result generated by the algorithm, each module having associated a metric of resource utilization by the module, the method including receiving a request for a machine learning task; and selecting a module from the plurality of modules for the task based on the metric associated with the module.

Embodiments of the disclosure relate to methods, apparatus and systems for authentication of a user. The described embodiments relate to obtaining ear biometric data for a user to be authenticated. The ear biometric data comprises one or more features characteristic of the user's ear canal and an associated fit metric indicative of a positioning of a personal audio device relative to the user's ear canal, the personal audio device comprising a transducer for application of acoustic stimulus to the user's ear to obtain the ear biometric data. The user may be identified as a particular authorised user based on one or more features and the associated fit metric.

An artificial intelligence based method and apparatus for processing information. A specific embodiment of the method includes: acquiring search click information recorded within a predetermined time period; generating a candidate entry set by selecting, from the search click information, entries having click volumes exceeding a click volume threshold within a preset unit time period; forming, for each candidate entry in the candidate entry set, a click volume sequence according to a chronological order of each of the click volumes corresponding to the candidate entry in the predetermined time period; determining, based on click volume sequences, categories of the candidate entries respectively corresponding to click volume sequences; and determining candidate entries having the categories being a preset category as points of interest to generate a set of points of interest.

Implementations of the subject matter described herein relate to object detection based on deep neural network. With a given input image, it is desired to determine a class and a boundary of one or more objects within the input image. Specifically, a plurality of channel groups is generated from a feature map of an image, the image including at least a region corresponding to a first grid. A target feature map is extracted from at least one of the plurality of channel groups associated with a cell of the first grid. Information related to an object within the region is determined based on the target feature map. The information related to the object may be a class and/or a boundary of the object.

A power electronic circuit fault diagnosis method based on Extremely randomized trees (ET) and Stack Sparse auto-encoder (SSAE) algorithm includes the following. First, collect the fault signal and extract fault features. Then, reduce the dimensionality of fault features by calculating the importance value of all features using ET algorithm. A proportion of the features to be eliminated is determined, and a new feature set is obtained according the value of importance. Further extraction of fault features is carried by using SSAE algorithm, and hidden layer features of the last sparse auto-encoder are obtained as fault features after dimensionality reduction. Finally, the fault samples in a training set and a test set are input to the classifier for training to obtain a trained classifier. And mode identification, wherein the fault of the power electronic circuit is identified and located by the training classifier.

In one embodiment, a method includes identifying datasets associated with a party and identifying one or more decentralized identifiers (DIDs) associated with the datasets. The method also includes generating an aggregated dataset associated with the DIDs and generating a training dataset associated with the aggregated dataset. The method further includes using one or more machine learning algorithms to recognize patterns within the training dataset.

Neural network-based categorization can be improved by incorporating graph neural networks that operate on a graph representing the taxonomy of the categories into which a given input is to be categorized by the neural network based-categorization. The output of a graph neural network, operating on a graph representing the taxonomy of categories, can be combined with the output of a neural network operating upon the input to be categorized, such as through an interaction of multidimensional output data, such as a dot product of output vectors. In such a manner, information conveying the explicit relationships between categories, as defined by the taxonomy, can be incorporated into the categorization. To recapture information, incorporate new information, or reemphasize information a second neural network can also operate upon the input to be categorized, with the output of such a second neural network being merged with the output of the interaction.

A system for selection and configuration of an automated robotic process includes a media input module structured to receive at least one functional media, a media analysis module structured to analyze the at least one functional media and identify an action parameter; and a solution selection module structured to select at least one component of an AI solution for use in an automated robotic process, wherein the selection is based, at least in part, on the action parameter.

A three-dimensional data encoding method includes: determining a total number of layers; when the total number of the layers is greater than 1, selecting, from pieces of attribute information of three-dimensional points, attribute information of a three-dimensional point based on a sampling period according to a data order of the pieces of attribute information of the three-dimensional points, and classifying the pieces of attribute information of the three-dimensional points into layers by assigning a first layer to pieces of attribute information of three-dimensional points selected and a second layer to pieces of attribute information of three-dimensional points non-selected; encoding the pieces of attribute information of the three-dimensional points for each of the layers; and generating a bitstream including the pieces of attribute information encoded, layer-number information, and sampling period information. When the total number of the layers is 1, a bitstream not including the sampling period information is generated.

A three-dimensional data encoding method includes: determining a total number of layers; when the total number of the layers is greater than 1, selecting, from pieces of attribute information of three-dimensional points, attribute information of a three-dimensional point based on a sampling period according to a data order of the pieces of attribute information of the three-dimensional points, and classifying the pieces of attribute information of the three-dimensional points into layers by assigning a first layer to pieces of attribute information of three-dimensional points selected and a second layer to pieces of attribute information of three-dimensional points non-selected; encoding the pieces of attribute information of the three-dimensional points for each of the layers; and generating a bitstream including the pieces of attribute information encoded, layer-number information, and sampling period information. When the total number of the layers is 1, a bitstream not including the sampling period information is generated.