A lightweight, non-intrusive memory anomaly detector has been designed that focuses on time sub-windows in the time-series data for selected memory related metrics that can efficiently be collected by probes or agents without being intrusive with the virtual machines (VMs) being monitored. In addition, the memory anomaly detector extracts features from those sub-windows of correlated features to present a smaller input vector to two classifiers: a fuzzy rule-based classifier and an artificial neural network. This allows the memory anomaly detector to be lightweight because it is less computationally expensive to run a smaller artificial neural network. The fuzzy rule-based classifier applies fuzzy rules to the input vector and provides classification labels, which are used to train an artificial neural network (ANN). After being trained, the trained ANN is refined with supervised feedback and presents its output of classification probabilities for application performance analysis.

Example techniques may be implemented as a method, a system or more non-transitory machine-readable media storing instructions that are executable by one or more processing devices, Operations performed by the example techniques include obtaining data representing results of tests executed by one or more test instruments on an initial set of devices under test (DUTs) in a test system; and using the data to train a machine learning model. The machine learning model is for predicting which of the tests will produce failing results for a different set of DUTs. DUTs in the different set have one or more features in common with DUTs in the initial set.

One embodiment provides a method and a system for diagnosing faults in a physical system. During operation, the system can create a fault-augmented model of the physical system by considering various potential faults, and it can generate a machine-learning model to predict an operation mode of the physical system using the outputs of the physical system. A respective operation mode corresponds to normal operation or a potential fault in the physical system. The system can generate a plurality of training samples based on the fault-augmented model, use the training samples to train the machine-learning model to learn a sequence of inputs and model parameters that minimizes an uncertainty of the predicted operation mode, and then apply the learned sequence of inputs and the trained machine-learning model on the physical system to determine the operation mode of the physical system.

The subject technology detects anomalies in media campaign configuration settings. The anomaly detection system may leverage one or more deep learning models to detect anomalies and identify particular configuration settings that contribute to the detected anomalies. In various embodiments, two or more of the deep learning models may be combined into an ensemble model that boosts the accuracy of anomaly predictions made by the anomaly detection system. The anomaly detection system may review the configuration settings of media campaigns during the configuration process and before the media campaigns run on a publication system in order to reduce the amount of unsuccessful campaigns and minimize the amount of wasted resources spent on running campaigns that have a low likelihood of achieving user defined goals.

Disclosed are a method for automatically diagnosing and controlling an apparatus in an intelligent building and relevant device. The method includes: performing, based on historical data of working parameters of multiple apparatuses, an abnormal diagnosis on received real-time data of the working parameters; determining an abnormal apparatus; selecting a neural network predictive control model corresponding to the abnormal apparatus; selecting one piece of non-abnormal data which has a same parameter type as that of the abnormal data and is close to the current abnormal data in time as a predictive control target, and determining a predictive control data that can cause an output matching the predictive control target; and controlling the abnormal apparatus according to the predictive control data. The automatic diagnosis and automatic control of an apparatus in an intelligent building are realized, meanwhile the safe and efficient operation of all apparatuses in an intelligent building is ensured.

An apparatus includes a processor configured to execute instructions, and a memory storing the instructions, which when executed by the processor configure the processor to generate system error prediction data using an error prediction neural network provided with one of a plurality of log data sequences generated by pre-processing a plurality of log data pieces of component log data of a system. The system error prediction data comprises information of a plurality of system errors occurring at a plurality of respective timepoints.

A Deep Belief Network (DBN) feature extraction-based analogue circuit fault diagnosis method comprises the following steps: a time-domain response signal of a tested analogue circuit is acquired, where the acquired time-domain response signal is an output voltage signal of the tested analogue circuit; DBN-based feature extraction is performed on the acquired voltage signal, wherein learning rates of restricted Boltzmann machines in a DBN are optimized and acquired by virtue of a quantum-behaved particle swarm optimization (QPSO); a support vector machine (SVM)-based fault diagnosis model is constructed, wherein a penalty factor and a width factor of an SVM are optimized and acquired by virtue of the QPSO; and feature data of test data are input into the SVM-based fault diagnosis model, and a fault diagnosis result is output, where the feature data of the test data is generated by performing the DBN-based feature extraction on the test data.

Methods, system, and computer readable medium are presented for predicting defects using a machine learning component based on a generated signature. A trained machine learning component that has been trained with historic data that represents a series of events that occurred within a plurality of heterogeneous systems over a plurality of periods of change for the heterogeneous systems can be received. A base signature for a first heterogeneous system that includes a first mix of modules can be compared to a current signature for the first heterogeneous system to identify one or more irregularities. The trained machine learning component can predict one or more defects for the first heterogeneous system based on the identified irregularity.

A computer implemented method includes receiving telemetry data corresponding to capacity health of nodes in a cloud based computing system. The received telemetry data is processed via a prediction engine to provide predictions of capacity health at multiple dimensions of the cloud based computing system. Node recoverability information is received and node recovery execution is initiated as a function of the representations of capacity health and node recoverability information.

A deep learning fault diagnosis method includes the following steps: a fault diagnosis data set X is processed based on sliding window processing, to obtain a picture-like sample data set {tilde over (X)}, and obtain an attention matrix A of the picture-like sample data set {tilde over (X)}; and a 2D-CNN model is constructed to process the picture-like sample data set {tilde over (X)} to obtain a corresponding feature map F, and in the meantime, the feature map F is processed based on channel-oriented average pooling and channel-oriented maximum pooling to obtain an output P.sub.1 of the average pooling and an output P.sub.2 of the maximum pooling, and a weight matrix W is obtained based on the attention matrix A, the output P.sub.1 of the average pooling, and the output P.sub.2 of the maximum pooling, so that an output of the model is a feature map {tilde over (F)} based on an attention mechanism, where {tilde over (F)}=WF.