Abstract: Aspects described herein may allow for the application of stochastic gradient boosting techniques to the training of deep neural networks by disallowing gradient back propagation from examples that are correctly classified by the neural network model while still keeping correctly classified examples in the gradient averaging. Removing the gradient contribution from correctly classified examples may regularize the deep neural network and prevent the model from overfitting. Further aspects described herein may provide for scheduled boosting during the training of the deep neural network model conditioned on a mini-batch accuracy and/or a number of training iterations. The model training process may start un-boosted, using maximum likelihood objectives or another first loss function.

Abstract: A device receives historical data associated with multiple cloud computing environments, trains one or more machine learning models, with the historical data, to generate trained machine learning models that generate outputs, and trains a model with the outputs to generate a trained model. The device receives particular data, associated with a cloud computing environment, that includes data identifying usage of resources associated with the cloud computing environment, and processes the particular data, with the trained machine learning models, to generate anomaly scores indicating anomalous usage of the resources associated with the cloud computing environment. The device processes the one or more anomaly scores, with the trained model, to generate a final anomaly score indicating anomalous usage of at least one of the resources associated with the cloud computing environment, and performs one or more actions based on the final anomaly score.

Abstract: Training query intents are allocated for multiple training entities into training time intervals in a time series based on a corresponding query intent time for each training query intent. Training performance results for the multiple training entities are allocated into the training time intervals in the time series based on a corresponding performance time of each training performance result. A machine learning model for a training milestone of the time series is trained based on the training query intents allocated to a training time interval prior to the training milestone and the training performance results allocated to a training time interval after the training milestone. Target performance for the target entity for an interval after a target milestone in the time series is predicted by inputting to the trained machine learning model target query intents allocated to the target entity in a target time interval before the target milestone.

Abstract: An inspection apparatus of the present disclosure includes: a machine learning device that performs machine learning on a basis of state data acquired from an inspection target and label data indicating an inspection result related to the inspection target to generate a learning model; a learning model evaluation index calculation unit that calculates a learning model evaluation index related to the learning model generated by the machine learning device as an evaluation index to be used to evaluate the learning model; an inspection index acquisition unit that acquires an inspection index to be used in the inspection; and a learning model selection unit that displays the learning model evaluation index and the inspection index so as to be comparable with each other regarding the learning model generated by the machine learning device, receives selection of the learning model by an operator, and outputs a result of the selection.

Abstract: A classification apparatus includes: an encoding module that includes an element classification part that extracts a feature of input data and outputs classification information based on an element classification model stored in a first storage unit; an integration module that includes an element estimation part that receives the classification information and converts the classification information to a collation vector based on an element estimation model stored in a second storage unit; and a determination module that includes a determination part that determines a group to which the collation vector belongs by collating the collation vector with a representative vector of an individual group stored as a semantic model in a third storage unit and outputs a group ID of the group as a classification result.

Abstract: The disclosure provides a method, a system and a storage medium for predicting power load probability density based on deep learning. The method comprises: S101, collecting power load data of a user, meteorological data and air quality data in a preset historical time period, and dividing the collected data into a training set and a test set; S102, determining a deep learning model for predicting power load; S103, inputting the test set into the deep learning model for predicting power load, and obtaining power load prediction data of the user at different quantile points in a third time interval; S104, performing kernel density estimation and obtaining a probability density curve of the power load of the user in the third time interval.

Abstract: Training query intents are allocated for multiple training entities into training time intervals in a time series based on a corresponding query intent time for each training query intent. Training performance results for the multiple training entities are allocated into the training time intervals in the time series based on a corresponding performance time of each training performance result. A machine learning model for a training milestone of the time series is trained based on the training query intents allocated to a training time interval prior to the training milestone and the training performance results allocated to a training time interval after the training milestone. Target performance for the target entity for an interval after a target milestone in the time series is predicted by inputting to the trained machine learning model target query intents allocated to the target entity in a target time interval before the target milestone.