Abstract: A system and method for classification. In some embodiments, the method includes forming a first training dataset and a second training dataset from a labeled input dataset; training a first classifier with the first training dataset; training a variational auto encoder with the second training dataset, the variational auto encoder comprising an encoder and a decoder; generating a third dataset, by feeding pseudorandom vectors into the decoder; labeling the third dataset, using the first classifier, to form a third training dataset; forming a fourth training dataset based on the third dataset; and training a second classifier with the fourth training dataset.

Abstract: Systems, methods, articles of manufacture, and computer program products to: train a prediction model using a machine learning process, the prediction model configured to estimate whether further application of a hyperparameter tuning technique will cause an improvement in at least one of the hyperparameters; select the hyperparameters using the tuning technique; apply the prediction model to determine if further adjustment of the hyperparameters is likely to improve the success metric; and terminate the tuning technique when: accuracy of the prediction model in predicting improvement in a hyperparameter is above a predetermined accuracy threshold, and the prediction model predicts that further application of the tuning technique will not result in an improvement to the hyperparameter; or the accuracy of the prediction model in predicting improvement in the parameter is below the predetermined accuracy threshold, and an accuracy of hyperparameter adjustment is determined to be below a predetermined adjustment

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for selecting a neural network to perform a particular machine learning task while satisfying a set of constraints.

Abstract: A hybrid knowledge representation is searched for a machine learning component corresponding to a search query. The hybrid knowledge representation may be structured as nodes representing machine learning workflow components and edges (e.g., links) connecting the nodes based on relationships between the nodes. Responsive to finding the machine learning component in the hybrid knowledge representation, the machine learning component is returned. Responsive to not finding the machine learning component in the hybrid knowledge representation, the hybrid knowledge representation is searched for machine learning model fragments associated with building the machine learning component, generating a new machine learning component by combining the machine learning model fragments and returning the new machine learning component.

Abstract: A processing device can establish a vector-trained, deep learning model to produce software dependency recommendations. The processing device can build a list of software dependencies and corresponding metatags for each of the software dependencies, and generate a probability distribution from the list. The processing device can sample the probability distribution to produce a latent vector space that includes representative vectors for the software dependencies. The processing device can train a hybrid deep learning model to produce software dependency recommendations using the latent vector space as well as collaborative data for the software dependencies.

Abstract: In disclosed approaches of neural network processing, a host computer system copies an input data matrix from host memory to a shared memory for performing neural network operations of a first layer of a neural network by a neural network accelerator. The host instructs the neural network accelerator to perform neural network operations of each layer of the neural network beginning with the input data matrix. The neural network accelerator performs neural network operations of each layer in response to the instruction from the host. The host waits until the neural network accelerator signals completion of performing neural network operations of layer i before instructing the neural network accelerator to commence performing neural network operations of layer i+1, for i?1. The host instructs the neural network accelerator to use a results data matrix in the shared memory from layer i as an input data matrix for layer i+1 for i?1.

Abstract: Systems and techniques for neuromorphic accelerator multitasking are described herein. A neuron address translation unit (NATU) may receive a spike message. Here, the spike message includes a physical neuron identifier (PNID) of a neuron causing the spike. The NATU may then translate the PNID into a network identifier (NID) and a local neuron identifier (LNID). The NATU locates synapse data based on the NID and communicates the synapse data and the LNID to an axon processor.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for selecting a neural network to perform a particular machine learning task while satisfying a set of constraints.

Abstract: A non-transitory computer-readable recording medium stores therein a program for causing a computer to execute a process for, in repeatedly training a given training model, repeatedly training the training model a given number of times by using a numerical value of a floating-point number, the numerical value being a parameter of the training model or training data of the training model, or any combination thereof; and, after the training by using the numerical value of the floating-point number, repeatedly training the training model by using a numerical value of a fixed-point number corresponding to a numerical value of the floating-point number obtained by the training.

Abstract: The disclosure generally describes methods, software, and systems, including a method for machine learning anomaly detection for a set of assets. Assets are analyzed using anomaly-detection analysis and a set of anomaly-detection rules. Each asset is associated with correlated records comprising characteristics of the particular asset and characteristic of non-asset-specific signals. Each anomaly-detection rule is associated with conditions determined to be indicative of a potential anomaly. At least a subset of the assets are provided for presentation in a user interface. Each asset is identified as being in a potential anomalous or non-anomalous state based on the anomaly-detection analysis. Input is received from a user identifying at least one asset as anomalous as a non-anomalous asset. Based on the received input, at least one anomaly-detection rule is modified that was applied to identify the asset as anomalous. The modified rule is stored for future analyses.

Abstract: At least one neural network accelerator performs operations of a first subset of layers of a neural network on an input data set, generates an intermediate data set, and stores the intermediate data set in a shared memory queue in a shared memory. A first processor element of a host computer system provides input data to the neural network accelerator and signals the neural network accelerator to perform the operations of the first subset of layers of the neural network on the input data set. A second processor element of the host computer system reads the intermediate data set from the shared memory queue, performs operations of a second subset of layers of the neural network on the intermediate data set, and generates an output data set while the neural network accelerator is performing the operations of the first subset of layers of the neural network on another input data set.

Abstract: In an example, for each of one or more terms in a text document, one or more entities to which the term potentially maps are identified. The text document includes at least one ambiguous term. One or more features are extracted from the text document. An attention model is applied to the text document based on the extracted one or more features, resulting in an attention weight being applied to each of the one or more terms in the text document. The one or more terms are encoded based on the attention weights. Each of one or more ambiguous terms is classified based on the encoded terms, the classification assigning a value to each different entity that each ambiguous term potentially maps to. A minimum entropy loss function is evaluated using the classification, and results are back-propagated to the attention model.

Abstract: Electric Grid Analytics Learning Machine, EGALM, is a machine learning based, “brutally empirical” analysis system for use in all energy operations. EGALM is applicable to all aspects of the electricity operations from power plants to homes and businesses. EGALM is a data-centric, computational learning and predictive analysis system that uses open source algorithms and unique techniques applicable to all electricity operations in the United States and other foreign countries.

Abstract: The subject disclosure relates to employing grouping and selection components to facilitate a grouping of failure data associated with oil and gas exploration equipment into one or more equipment failure type groups. In an example, a method comprises grouping, by a system operatively coupled to a processor, training data of a set of equipment failure data into one or more failure type groups based on one or more determined failure criteria, wherein the one or more failure type groups represent equipment failure classifications associated with energy exploration processes; and selecting, by the system, first ungrouped data from the set of equipment failure data based on a level of similarity between the first ungrouped data and the training data.

Abstract: The subject disclosure relates to employing grouping and selection components to facilitate a grouping of failure data associated with oil and gas exploration equipment into one or more equipment failure type groups. In an example, a method comprises grouping, by a system operatively coupled to a processor, training data of a set of equipment failure data into one or more failure type groups based on one or more determined failure criteria, wherein the one or more failure type groups represent equipment failure classifications associated with energy exploration processes; and selecting, by the system, first ungrouped data from the set of equipment failure data based on a level of similarity between the first ungrouped data and the training data.

Abstract: Methods, apparatus, and machine-readable mediums are described for selecting a training set from a larger data set. Samples are divided into a training set and a validation set. Each set meets one or more conditions. For each class to be modeled, multiple training sets are created. Models are trained on each of the multiple training sets. A size of samples for each class is determined based upon the trained models. A training data set that includes a number of samples based upon the determined size of samples is created.

Abstract: A system may include a decentralized communication network and multiple processing devices on the network. Each processing device may have an artificial intelligence (AI) chip, the device may be configured to generate an AI model, determine the performance value of the AI model on the AI chip, receive a chain from the network where the chain contains a performance measure. If the performance value of the AI model is better than the performance measure, then the processing device may broadcast the AI model to the network for verification. If the AI model is verified by the network, the device may update the chain with the performance value so that the chain can be shared by the multiple processing devices on the network. Any processing device on the network may also verify an AI model broadcasted by any other device. Methods for generating the AI model are also provided.

Abstract: A system may include a decentralized communication network and multiple processing devices on the network. Each processing device may have an artificial intelligence (AI) chip, the device may be configured to generate an AI model, determine the performance value of the AI model on the AI chip, receive a chain from the network where the chain contains a performance measure. If the performance value of the AI model is better than the performance measure, then the processing device may broadcast the AI model to the network for verification. If the AI model is verified by the network, the device may update the chain with the performance value so that the chain can be shared by the multiple processing devices on the network. Any processing device on the network may also verify an AI model broadcasted by any other device. Methods for generating the AI model are also provided.