Abstract: Embodiments of the present disclosure provide a method and an apparatus for training a model, an electronic device, and a medium. This method includes: generating a first group of features and a second group of features respectively from a first sample set and a second sample set based on the model, wherein the first sample set is of a first category, and the second sample set is of a second category different from the first category; generating a first similarity matrix for the first sample set and the second sample set based on the first group of features and the second group of features; determining a first loss for the first sample set and the second sample set based on the first similarity matrix; and updating the model based on the first loss.

Abstract: Embodiments of the present disclosure include a method, an electronic device, and a computer program product for training a failure analysis model. In a method for training a failure analysis model in an illustrative embodiment, at least one set of log files including multiple preprocessed log files is obtained, the at least one set of log files including a marked failure cause of a storage system, and preprocessed log files in the multiple preprocessed log files including one or more potential failure causes of the storage system and scores associated with the potential failure causes; a failure cause of the storage system is predicted according to a failure analysis model and based on the potential failure causes and the scores in the multiple preprocessed log files; and parameters of the failure analysis model are updated based on a probability that the predicted failure cause is the marked failure cause.

Abstract: Embodiments of the present disclosure relate to a computer-implemented method, a device, and a computer program product. The method includes extracting respective themes of a set of documents with release time within a first period; determining respective semantic information of the themes and frequencies of the themes appearing in the set of documents; and determining the number of documents associated with the themes within a second period according to a prediction model and based on the semantic information and frequencies of the themes. The second period is after the first period. Embodiments of the present disclosure can better predict the tendency of the themes appearing in the future based on the semantic information and frequencies of the themes.

Abstract: Implementations of the present disclosure relate to a method, an electronic device, and a computer program product for managing an inference process. Here, the inference process is implemented based on a machine learning model. A method includes: determining, based on a computational graph defining the machine learning model, dependency relationships between a set of functions for implementing the inference process; acquiring, in at least one edge device located in an edge computing network, a set of computing units available to execute the inference process; selecting at least one computing unit for executing the set of functions from the set of computing units; and causing the at least one computing unit to execute the set of functions based on the dependency relationships. With example implementations of the present disclosure, the inference process is implemented by making use of a variety of computing units in the edge computing network, thereby improving performance.

Abstract: Embodiments of the present disclosure include a method, an electronic device, and a computer program product for training a failure analysis model. In a method for training a failure analysis model in an illustrative embodiment, at least one set of log files including multiple preprocessed log files is obtained, the at least one set of log files including a marked failure cause of a storage system, and preprocessed log files in the multiple preprocessed log files including one or more potential failure causes of the storage system and scores associated with the potential failure causes; a failure cause of the storage system is predicted according to a failure analysis model and based on the potential failure causes and the scores in the multiple preprocessed log files; and parameters of the failure analysis model are updated based on a probability that the predicted failure cause is the marked failure cause.

Abstract: A method for graph data processing comprises obtaining graph data which includes a plurality of nodes and data corresponding to the plurality of nodes respectively; classifying the plurality of nodes into at least one category of a plurality of categories, wherein the plurality of categories are associated with a plurality of node relationship patterns; determining, from a plurality of candidate parameter value sets of a graph convolutional network (GCN) model, parameter value subsets respectively matching at least one category, wherein the plurality of candidate parameter value sets are determined by training the GCN model respectively for the plurality of node relationship patterns; and using the parameter value subsets respectively matching the at least one category to respectively perform a graph convolution operation in the GCN model on data corresponding to the nodes classified into the at least one category to obtain a processing result for the graph data.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for analyzing samples. The method includes acquiring a set of feature representations associated with a set of samples. The set of samples illustratively have classification information for indicating classifications of the set of samples. The method further includes adjusting the set of feature representations so that distances between feature representations of samples corresponding to the same classification are less than a first distance threshold. The method further includes training a classification model based on the adjusted set of feature representations and the classification information. The classification model is illustratively configured to receive an input sample and determine a classification of the input sample. In this manner, a relatively accurate classification model can be trained using a small number of samples, thereby reducing computation time and required computation capacity.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for image recognition. In some embodiments, characterization information for a first reference image in a reference image set is generated in an image recognition engine by using a Gaussian mixture model. First reference label information for the first reference image is generated based on the characterization information for the first reference image, the first reference label information being associated with a category of a first object in the first reference image. The image recognition engine is updated by determining the accuracy of the first reference label information for the first reference image. In this way, good characterization of images and generation of reference label information for the images can be achieved, thus both improving the robustness of the generated reference label information and significantly improving the accuracy of image recognition.

Abstract: A method for graph data processing comprises obtaining graph data which includes a plurality of nodes and data corresponding to the plurality of nodes respectively; classifying the plurality of nodes into at least one category of a plurality of categories, wherein the plurality of categories are associated with a plurality of node relationship patterns; determining, from a plurality of candidate parameter value sets of a graph convolutional network (GCN) model, parameter value subsets respectively matching at least one category, wherein the plurality of candidate parameter value sets are determined by training the GCN model respectively for the plurality of node relationship patterns; and using the parameter value subsets respectively matching the at least one category to respectively perform a graph convolution operation in the GCN model on data corresponding to the nodes classified into the at least one category to obtain a processing result for the graph data.

Abstract: Embodiments of the present disclosure relate to a computer-implemented method, a device, and a computer program product. The method includes: determining, based on a set of sample features extracted from an input sample by a feature extraction model, a confidence level of the input sample and a similarity degree among the set of sample features; determining a first loss based on the confidence level, the set of sample features, and label information for the input sample, the first loss being related to the quality of the label information; determining a second loss based on the similarity degree among the set of sample features, the second loss being related to the quality of the set of sample features; and training the feature extraction model based on the first loss and the second loss. Embodiments of the present disclosure determine the confidence level of the input sample, thereby optimizing the feature extraction model.

Abstract: The present disclosure relates to a method, a device, and a program product for processing sample data in an Internet of Things environment. A method in one embodiment includes: receiving features of the sample data from an encoder deployed in a remote device in the Internet of Things environment; acquiring a category probability corresponding to the sample data based on a classifier deployed in a local device in the Internet of Things environment and the features; and classifying the sample data to a predetermined category in response to determining that the category probability satisfies a first threshold condition. Further, a corresponding device and a corresponding program product are provided. With example implementations of the present disclosure, computing resources of devices in an Internet of Things environment can be fully utilized to process sample data.

Abstract: A method in an illustrative embodiment includes determining a first set of distilled samples from a first set of samples based on a characteristic distribution of the first set of samples, the first set of samples being associated with a first set of classifications. The method also includes acquiring a first set of characteristic representations associated with the first set of distilled samples. The method also includes adjusting the first set of characteristic representations so that a distance between characteristic representations associated with the same classification is less than a predetermined threshold.

Abstract: Embodiments of the present disclosure relate to a model optimization method, an electronic device, and a computer program product. This method includes: determining an initial learning rate combination for a deep learning model, wherein the initial learning rate combination includes a plurality of learning rates, each learning rate being determined for one of a plurality of layers of the deep learning model, and the plurality of learning rates including static learning rates and dynamic learning rates; and adjusting the initial learning rate combination to obtain a target learning rate combination, wherein an accuracy rate achieved when the target learning rate combination is used to train the deep learning model is higher than or equal to a first threshold accuracy rate. With the technical solution of the present disclosure, a deep learning model can be optimized by setting learning rates for each layer of the deep learning model.

Abstract: Embodiments disclosed herein include a method, an electronic device, and a computer program product for data processing. The method includes determining a first set of feature vectors representing samples in a data set. The method also includes generating a second set of feature vectors by performing a first transformation on the first set of feature vectors, wherein distribution skewness of the second set of feature vectors in a feature space is smaller than that of the first set of feature vectors. The method also includes generating a third set of feature vectors by performing a second transformation on the second set of feature vectors, wherein the third set of feature vectors and the second set of feature vectors have different distances between vectors. The method also includes selecting target samples as representatives from the samples based on a distribution of the third set of feature vectors in the feature space.

Abstract: The present disclosure relates to a method, a device, and a computer program product for model updating. The method includes: acquiring a first image set and first annotation information, wherein the first annotation information indicates whether a corresponding image in the first image set includes a target object; updating a first version of an object verification model using the first image set and the first annotation information to obtain a second version, wherein the first version of the object verification model has been deployed to determine whether an input image includes the target object; determining the accuracy of the second version of the object verification model; and updating, if it is determined that the accuracy is lower than a preset accuracy threshold, the second version of the object verification model using a second image set and second annotation information to obtain a third version of the object verification model.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for processing information. According to an example embodiment, the method includes: acquiring a service request record set, each service request record in the service request record set relating to a problem encountered by a user when the user is provided with a service and a solution to the problem; constructing a language model based on a first subset in the service request record set and an initial model, the initial model being trained using a predetermined corpus and configured to determine vector representations of words and sentences in the corpus; and constructing a classification model based on a second subset in the service request record set and the language model, the classification model being capable of determining a solution to a pending problem, and the first subset being different from the second subset.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for data distillation. The method includes: training an input data set by using a machine learning training process to establish a training model of the input data set; extracting multiple weights from the training model of the input data set, wherein the multiple weights contain information indicating the input data set, and the multiple weights are orthogonal to each other; and retraining the training model by using the multiple weights for generating a reconstructed data set. The embodiments of the present disclosure can greatly reduce the data storage cost of a data storage system and maintain the performance of the data storage system.

Abstract: Embodiments of the present disclosure relate to an article processing method, electronic device, and computer program product. The method includes: determining, based on content of a target article, a target article vector associated with the target article; acquiring a reference article vector set associated with a reference article set; and determining, based on a distance in an article vector space between the target article vector and a reference article vector in the reference article vector set, a reference article vector associated with the target article vector in the reference article vector set as an association article vector. By using the technical solution of the present disclosure, an association article associated with a target article can be accurately provided based on the target article selected by a user, so that reports on the target article and its association articles can be further provided to the user for analysis and selection.

Abstract: Techniques for constructing and otherwise managing knowledge graphs in information processing system environments are disclosed. For example, a method comprises the following steps. The method collects data from a plurality of data sources. The method extracts structured data and unstructured data from the collected data, wherein unstructured data is extracted using an unsupervised machine learning process. The method forms a plurality of sub-graph structures comprising a sub-graph structure for each of the data sources based on at least a portion of the extracted structured data and unstructured data. The method combines the plurality of sub-graph structures to form a combined graph structure representing the collected data from the plurality of data sources. The resulting combined graph structure is a comprehensive knowledge graph.

Abstract: An apparatus comprises a processing device configured to receive a service request associated with a given asset, to obtain a log file associated with the given asset, to split the log file into log segments, to generate sets of log pattern identifiers for the log segments, and to determine risk scores for the log segments utilizing a machine learning model that takes as input the sets of log pattern identifiers and provides as output information characterizing risk of the log segments. The processing device is also configured to identify critical areas of the log file based at least in part on the determined risk scores, a given critical area comprising a sequence of log segments having determined risk scores above a designated risk score threshold. The processing device is further configured to analyze the identified critical areas to determine remedial actions to be applied for resolving the service request.