Abstract: Techniques for implementing a semi-supervised framework for purpose-oriented anomaly detection are provided. In one technique, a data item in inputted into an unsupervised anomaly detection model, which generates first output. Based on the first output, it is determined whether the data item represents an anomaly. In response to determining that the data item represents an anomaly, the data item is inputted into a supervised classification model, which generates second output that indicates whether the data item is unknown. In response to determining that the data item is unknown, a training instance is generated based on the data item. The supervised classification model is updated based on the training instance.

Abstract: Herein are feature extraction mechanisms that receive parsed log messages as inputs and transform them into numerical feature vectors for machine learning models (MLMs). In an embodiment, a computer extracts fields from a log message. Each field specifies a name, a text value, and a type. For each field, a field transformer for the field is dynamically selected based the field's name and/or the field's type. The field transformer converts the field's text value into a value of the field's type. A feature encoder for the value of the field's type is dynamically selected based on the field's type and/or a range of the field's values that occur in a training corpus of an MLM. From the feature encoder, an encoding of the value of the field's typed is stored into a feature vector. Based on the MLM and the feature vector, the log message is detected as anomalous.

Abstract: Herein are machine learning techniques that adjust reconstruction loss of a reconstructive model such as an autoencoder based on importances of values of features. In an embodiment and before, during, or after training, the reconstructive model that more or less accurately reconstructs its input, a computer measures, for each distinct value of each feature, a respective importance that is not based on the reconstructive model. For example, importance may be based solely on a training corpus. For each feature during or after training, a respective original loss from the reconstructive model measures a difference between a value of the feature in an input and a reconstructed value of the feature generated by the reconstructive model. For each feature, the respective importance of the input value of the feature is applied to the respective original loss to generate a respective weighted loss. The weighted losses of the features of the input are collectively detected as anomalous or non-anomalous.

Abstract: Techniques are provided for sparse ensembling of unsupervised machine learning models. In an embodiment, the proposed architecture is composed of multiple unsupervised machine learning models that each produce a score as output and a gating network that analyzes the inputs and outputs of the unsupervised machine learning models to select an optimal ensemble of unsupervised machine learning models. The gating network is trained to choose a minimal number of the multiple unsupervised machine learning models whose scores are combined to create a final score that matches or closely resembles a final score that is computed using all the scores of the multiple unsupervised machine learning models.

Abstract: Techniques are provided herein for contextual embedding of features of operational logs or network traffic for anomaly detection based on sequence prediction. In an embodiment, a computer has a predictive recurrent neural network (RNN) that detects an anomalous network flow. In an embodiment, an RNN contextually transcodes sparse feature vectors that represent log messages into dense feature vectors that may be predictive or used to generate predictive vectors. In an embodiment, graph embedding improves feature embedding of log traces. In an embodiment, a computer detects and feature-encodes independent traces from related log messages. These techniques may detect malicious activity by anomaly analysis of context-aware feature embeddings of network packet flows, log messages, and/or log traces.

Abstract: Herein are feature extraction mechanisms that receive parsed log messages as inputs and transform them into numerical feature vectors for machine learning models (MLMs). In an embodiment, a computer extracts fields from a log message. Each field specifies a name, a text value, and a type. For each field, a field transformer for the field is dynamically selected based the field's name and/or the field's type. The field transformer converts the field's text value into a value of the field's type. A feature encoder for the value of the field's type is dynamically selected based on the field's type and/or a range of the field's values that occur in a training corpus of an MLM. From the feature encoder, an encoding of the value of the field's typed is stored into a feature vector. Based on the MLM and the feature vector, the log message is detected as anomalous or not.

Abstract: Techniques are described herein for estimating CPU, memory, and I/O utilization for a workload via out-of-band sensor readings using a machine learning model. The framework involves receiving sensor data associated with executing benchmark applications, obtaining ground truth utilization values for the benchmarks, preprocessing the training data to select a set of enhanced sequences, and using the enhanced sequences to train a random forest model to estimate CPU, memory, and I/O utilization given sensor monitoring data. Prior to the training phase, a machine learning model is trained using a set of predefined hyper-parameters. The trained models are used to generate estimations for CPU, memory, and I/O utilizations values. The utilization values are used with workload context information to assess the deployment and generate one or more recommendations for machine types that will best serve the workload in terms of system utilization.

Abstract: Herein are techniques for training a parser by categorizing and generalizing messages and abstracting message templates for parsing after training. In an embodiment, a computer generates a message signature based on a message sequence of tokens that were extracted from a training message. The message signature is matched to a cluster signature that represents messages of one of many clusters that have distinct signatures. The training message is added to the cluster. Based on a data type of the cluster signature, a value is extracted from a second message, such as a live message after training. Fuzzy signatures may be probabilistically matched to select a best matching cluster for a message. The value range of a token may be broadened or narrowed by adding or removing candidate data types, by adding or removing literals to a data type, and/or by promoting a narrow data type to a broader data type.

Abstract: Approaches herein relate to reconstructive models such as an autoencoder for anomaly detection. Herein are machine learning techniques that detect and suppress any feature that causes model decay by concept drift. In an embodiment in a production environment, a computer initializes an unsuppressed subset of features with a plurality of features that an already-trained reconstructive model can process. A respective reconstruction error of each feature of the unsuppressed subset of features is calculated. The computer detects that a respective moving average based on the reconstruction error of a particular feature of the unsuppressed subset of features exceeds a respective feature suppression threshold of the particular feature, which causes removal of the particular feature from the unsuppressed subset of features.

Abstract: Approaches herein relate to model decay of an anomaly detector due to concept drift. Herein are machine learning techniques for dynamically self-tuning an anomaly score threshold. In an embodiment in a production environment, a computer receives an item in a stream of items. A machine learning (ML) model hosted by the computer infers by calculation an anomaly score for the item. Whether the item is anomalous or not is decided based on the anomaly score and an adaptive anomaly threshold that dynamically fluctuates. A moving standard deviation of anomaly scores is adjusted based on a moving average of anomaly scores. The moving average of anomaly scores is then adjusted based on the anomaly score. The adaptive anomaly threshold is then adjusted based on the moving average of anomaly scores and the moving standard deviation of anomaly scores.

Abstract: Approaches herein relate to reconstructive models such as an autoencoder for anomaly detection. Herein are machine learning techniques that measure inference confidence based on reconstruction error trends. In an embodiment, a computer hosts a reconstructive model that encodes and decodes features. Based on that decoding, the following are automatically calculated: a respective reconstruction error of each feature, a respective moving average of reconstruction errors of each feature, an average of the moving averages of the reconstruction errors of all features, a standard deviation of the moving averages of the reconstruction errors of all features, and a confidence of decoding the features that is based on a ratio of the average of the moving averages of the reconstruction errors to the standard deviation of the moving averages of the reconstruction errors. The computer detects and indicates that a threshold exceeds the confidence of decoding, which may cause important automatic reactions herein.

Abstract: A multilayer perceptron herein contains an already-trained combined sequence of residual blocks that contains a semantic sequence of residual blocks and a contextual sequence of residual blocks. The semantic sequence of residual blocks contains a semantic sequence of layers of an autoencoder. The contextual sequence of residual blocks contains a contextual sequence of layers of a recurrent neural network. Each residual block of the combined sequence of residual blocks is used based on a respective survival probability. By the autoencoder and based on the using each residual block of the semantic sequence, a previous entry of a log is semantically encoded. By the recurrent neural network and based on the using each residual block of the contextual sequence, a next entry of the log is predicted. In an embodiment during training, survival probabilities are hyperparameters that are learned and used to probabilistically skip residual blocks such that the multilayer perceptron has stochastic depth.

Abstract: Techniques are provided herein for contextual embedding of features of operational logs or network traffic for anomaly detection based on sequence prediction. In an embodiment, a computer has a predictive recurrent neural network (RNN) that detects an anomalous network flow. In an embodiment, an RNN contextually transcodes sparse feature vectors that represent log messages into dense feature vectors that may be predictive or used to generate predictive vectors. In an embodiment, graph embedding improves feature embedding of log traces. In an embodiment, a computer detects and feature-encodes independent traces from related log messages. These techniques may detect malicious activity by anomaly analysis of context-aware feature embeddings of network packet flows, log messages, and/or log traces.

Abstract: Techniques are provided herein for contextual embedding of features of operational logs or network traffic for anomaly detection based on sequence prediction. In an embodiment, a computer has a predictive recurrent neural network (RNN) that detects an anomalous network flow. In an embodiment, an RNN contextually transcodes sparse feature vectors that represent log messages into dense feature vectors that may be predictive or used to generate predictive vectors. In an embodiment, graph embedding improves feature embedding of log traces. In an embodiment, a computer detects and feature-encodes independent traces from related log messages. These techniques may detect malicious activity by anomaly analysis of context-aware feature embeddings of network packet flows, log messages, and/or log traces.

Abstract: Herein are techniques for efficient and modular transcoding of message fields into features for inclusion within a feature vector. In an embodiment, a computer receives message signatures. Each signature has fields. Each field has a name and type. A feature map is generated that associates a field name and field type with transcoder(s). A message is received from a parser as field tuples. Each tuple has a type, name, and value of a field. Each tuple is processed as follows. The field name and field type of the tuple is used as a lookup key into the feature map to retrieve respective transcoder(s) that each generate a respective encoded feature from the field value of the tuple. An encoded feature from at least one relevant transcoder is written into a respective distinct location within a feature vector to encode the message. An inference is made based on the feature vector.

Abstract: Herein are techniques for analysis of data streams. In an embodiment, a computer associates each software actor with data streams. Each software actor has its own backlog queue of data to analyze. In response to receiving some stream content and based on the received stream content, data is distributed to some software actors. In response to determining that the data satisfies completeness criteria of a particular software actor, an indication of the data is appended onto the backlog queue of the particular software actor. The particular software actor is reset to an initial state by loading an execution snapshot of a previous initial execution of an embedded virtual machine. Based on the particular software actor, execution of the execution snapshot of the previous initial execution is resumed to dequeue and process the indication of the data from the backlog queue of the particular software actor to generate a result.

Abstract: Herein are techniques for training a parser by categorizing and generalizing messages and abstracting message templates for parsing after training. In an embodiment, a computer generates a message signature based on a message sequence of tokens that were extracted from a training message. The message signature is matched to a cluster signature that represents messages of one of many clusters that have distinct signatures. The training message is added to the cluster. Based on a data type of the cluster signature, a value is extracted from a second message, such as a live message after training. Fuzzy signatures may be probabilistically matched to select a best matching cluster for a message. The value range of a token may be broadened or narrowed by adding or removing candidate data types, by adding or removing literals to a data type, and/or by promoting a narrow data type to a broader data type.

Abstract: Techniques are described herein for estimating CPU, memory, and I/O utilization for a workload via out-of-band sensor readings using a machine learning model. The framework involves receiving sensor data associated with executing benchmark applications, obtaining ground truth utilization values for the benchmarks, preprocessing the training data to select a set of enhanced sequences, and using the enhanced sequences to train a random forest model to estimate CPU, memory, and I/O utilization given sensor monitoring data. Prior to the training phase, a machine learning model is trained using a set of predefined hyper-parameters. The trained models are used to generate estimations for CPU, memory, and I/O utilizations values. The utilization values are used with workload context information to assess the deployment and generate one or more recommendations for machine types that will best serve the workload in terms of system utilization.

Abstract: Herein are techniques for analysis of data streams. In an embodiment, a computer associates each software actor with data streams. Each software actor has its own backlog queue of data to analyze. In response to receiving some stream content and based on the received stream content, data is distributed to some software actors. In response to determining that the data satisfies completeness criteria of a particular software actor, an indication of the data is appended onto the backlog queue of the particular software actor. The particular software actor is reset to an initial state by loading an execution snapshot of a previous initial execution of an embedded virtual machine. Based on the particular software actor, execution of the execution snapshot of the previous initial execution is resumed to dequeue and process the indication of the data from the backlog queue of the particular software actor to generate a result.

Abstract: Techniques are provided herein for contextual embedding of features of operational logs or network traffic for anomaly detection based on sequence prediction. In an embodiment, a computer has a predictive recurrent neural network (RNN) that detects an anomalous network flow. In an embodiment, an RNN contextually transcodes sparse feature vectors that represent log messages into dense feature vectors that may be predictive or used to generate predictive vectors. In an embodiment, graph embedding improves feature embedding of log traces. In an embodiment, a computer detects and feature-encodes independent traces from related log messages. These techniques may detect malicious activity by anomaly analysis of context-aware feature embeddings of network packet flows, log messages, and/or log traces.