Abstract: A neural network device may include an input unit suitable for applying input signals to corresponding first lines, a calculating unit including memory elements cross-connected between the first lines and second lines, wherein the memory elements have respective weight values and generate product signals of input signals of corresponding first lines from among the plurality of first lines and weights to output the product signals to corresponding second lines from among the second lines, a drop-connect control unit including switches connected between the plurality of first lines and the plurality of memory elements, and suitable for randomly dropping a connection of an input signal applied to a corresponding memory element from among the plurality of memory elements, and an output unit connected to the plurality of second lines, and suitable for selectively activating signals of the plurality of second lines to apply the activated signals to the input unit and performing an output for the activated signals w

Abstract: Each energy value calculation circuit calculates an energy value, based on a sum total of values obtained by multiplying state values of a plurality of second neurons coupled with a first neuron by corresponding weighting values indicating coupling strengths, and updates the energy value, based on identification information of an updated neuron whose state is updated among the plurality of second neurons and a state value of the updated neuron. Each state transition determination circuit outputs, based on a second energy value and a noise value, a determination signal indicating a determination result of whether a change in a state value of the first neuron is possible. An updated neuron selection circuit selects, based on received determination signals, a first neuron a change in whose state value is possible and outputs identification information of the selected first neuron as identification information of the updated neuron.

Abstract: Methods, systems, and computer-readable storage media for receiving a vocabulary, the vocabulary including text data that is provided as at least a portion of raw data, the raw data being provided in a computer-readable file, associating each word in the vocabulary with a feature vector, providing a sentence embedding for each sentence of the vocabulary based on a plurality of feature vectors to provide a plurality of sentence embeddings, providing a reconstructed sentence embedding for each sentence embedding based on a weighted parameter matrix to provide a plurality of reconstructed sentence embeddings, and training the unsupervised neural attention model based on the sentence embeddings and the reconstructed sentence embeddings to provide a trained neural attention model, the trained neural attention model being used to automatically determine aspects from the vocabulary.

Abstract: Methods and systems for automated tuning of a service configuration are disclosed. An optimal configuration for a test computer is selected by performing one or more load tests using the test computer for each of a plurality of test configurations. The performance of a plurality of additional test computers configured with the optimal configuration is automatically determined by performing additional load tests using the additional test computers. A plurality of production computers are automatically configured with the optimal configuration if the performance of the additional test computers is improved with the optimal configuration.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using multi-task neural networks. One of the methods includes receiving a first network input and data identifying a first machine learning task to be performed on the first network input; selecting a path through the plurality of layers in a super neural network that is specific to the first machine learning task, the path specifying, for each of the layers, a proper subset of the modular neural networks in the layer that are designated as active when performing the first machine learning task; and causing the super neural network to process the first network input using (i) for each layer, the modular neural networks in the layer that are designated as active by the selected path and (ii) the set of one or more output layers corresponding to the identified first machine learning task.

Abstract: A method of configuring a System-on-Chip (SoC) to execute a Convolutional Neural Network (CNN) by (i) receiving scheduling schemes each specifying a sequence of operations executable by Processing Units (PUs) of the SoC; (ii) selecting, a scheduling scheme for a current layer of the CNN; (iii) determining a current state of memory for a storage location in the SoC allocated for storing feature map data from the CNN; (iv) selecting, from the plurality of scheduling schemes and dependent upon the scheduling scheme for the current layer of the CNN, a set of candidate scheduling schemes for a next layer of the CNN; and (v) selecting, from the set of candidate scheduling schemes dependent upon the determined current state of memory, a scheduling scheme for the next layer of the CNN.

Abstract: A network is crawled using a trained learning model to identify a set of secondary-source documents related to an event. A hub page from the set of secondary-source documents is identified that includes a link predicted to link to a new relevant secondary-source document. The new document is added to the set of secondary-source documents. Information is extracted from the set of secondary-source documents. Feedback is received indicative of a relevancy level for the extracted information as applied to the event. Each document is classified into one or more categories related to the event, based on the extracted information and the received feedback information. A learning model is trained based on the received feedback.

Abstract: A search apparatus receives an input target value, which indicates a condition to be set in a semiconductor processing apparatus or a result obtained by processing the semiconductor using the processing apparatus, a reference value of the condition inside a search area, and the result, wherein the reference value is indicated by the target value. A prediction model indicating a relation between the condition and the result based on a setting value of the condition inside the search area is generated and, a measured value of the result is obtained. A prediction value is acquired by assigning the target value to the prediction model. The prediction value is set to the reference value when it is determined that the prediction value is closer to the target value, and a prediction value satisfying an achievement condition is set when the prediction value satisfies the achievement condition of the target value.

Abstract: In an example embodiment, a machine learning algorithm is used to train a query-based deep semantic similarity neural network to output a query context vector in a vector space that includes both query context vectors and document context vectors. Both the query context vectors and document context vectors are clustered using a clustering algorithm. When an input search query is obtained, the input search query is also passed into the query-based deep semantic similarity neural network and its output document context vector assigned to a first cluster based on the clustering algorithm. Documents within the first cluster are then retrieved in response to the input search query.

Abstract: A compression coding apparatus for artificial neural network, including memory interface unit, instruction cache, controller unit and computing unit, wherein the computing unit is configured to perform corresponding operation to data from the memory interface unit according to instructions of controller unit; the computing unit mainly performs three steps operation: step one is to multiply input neuron by weight data; step two is to perform adder tree computing and add the weighted output neuron obtained in step one level-by-level via adder tree, or add bias to output neuron to get biased output neuron; step three is to perform activation function operation to get final output neuron. The present disclosure also provides a method for compression coding of multi-layer neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating an output sequence from an input sequence. In one aspect, one of the systems includes an encoder neural network configured to receive the input sequence and generate encoded representations of the network inputs, the encoder neural network comprising a sequence of one or more encoder subnetworks, each encoder subnetwork configured to receive a respective encoder subnetwork input for each of the input positions and to generate a respective subnetwork output for each of the input positions, and each encoder subnetwork comprising: an encoder self-attention sub-layer that is configured to receive the subnetwork input for each of the input positions and, for each particular input position in the input order: apply an attention mechanism over the encoder subnetwork inputs using one or more queries derived from the encoder subnetwork input at the particular input position.

Abstract: An event clustering system and associated methods include a feedback signalizer functor that responds to user interactions with already formed situations. The system and method then learns how to replicate the same situation when new alerts reoccur, or, creates similar situations. The feedback signalizer functor is a supervised machine learning approach to train a signalizer functor to reproduce a situation at varying degrees of precision.

Abstract: According to one embodiment, a computer program product for performing chemical textual analysis comprises a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, and where the program instructions are executable by a processor to cause the processor to perform a method comprising identifying a textual document, utilizing the processor, determining chemical data within the textual document, utilizing the processor, performing an analysis of the chemical data to identify a chemical pathway, utilizing the processor, and calculating a probability score for the chemical pathway, utilizing the processor.

Abstract: A system that includes one or more processors and one or more memories storing code that is executable by the one or more processors to: access a data log corresponding to a diet of a user over a predetermined period; determine a diet context for the user based at least in part on the data log; analyze the data log to determine a variance in the diet context; determine a potential medical issue in response to the variance; and generate predictive feedback for the user in response to the potential medical issue.

Abstract: Presented are relation inference methods and systems that use deep learning techniques for data mining documents to discover a relation between terms of interest in a given field covering a specific topic. For example, in the healthcare domain, various embodiments of the present disclosure provide for a relation inference system that mines large-scale medical documents in a free-text database to extract symptom and disease terms and generates relation information that aids in disease diagnosis. In embodiments, this is accomplished by training and using an RNN, such as an LSTM, a Gated Recurrent Unit (GRU), etc., that takes advantage of a term dictionary to examine co-occurrences of terms of interest within documents to discover correlations between the terms. The correlation may then be used to predict statistically most probable terms (e.g., a disease) related to a given search term (e.g., a symptom).

Abstract: Methods, systems, and computer storage media for implementing neural networks in fixed point arithmetic computing systems. In one aspect, a method includes the actions of receiving a request to process a neural network using a processing system that performs neural network computations using fixed point arithmetic; for each node of each layer of the neural network, determining a respective scaling value for the node from the respective set of floating point weight values for the node; and converting each floating point weight value of the node into a corresponding fixed point weight value using the respective scaling value for the node to generate a set of fixed point weight values for the node; and providing the sets of fixed point floating point weight values for the nodes to the processing system for use in processing inputs using the neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for performing multi-task learning. In one method a system obtains a respective set of training data for each of multiple machine learning tasks. For each of the machine learning tasks, the system configures a respective teacher machine learning model to perform the machine learning task by training the teacher machine learning model on the training data. The system trains a single student machine learning model to perform the multiple machine learning tasks using (i) the configured teacher machine learning models, and (ii) the obtained training data.

Abstract: A parallel convolutional neural network is provided. The CNN is implemented by a plurality of convolutional neural networks each on a respective processing node. Each CNN has a plurality of layers. A subset of the layers are interconnected between processing nodes such that activations are fed forward across nodes. The remaining subset is not so interconnected.

Abstract: The disclosed embodiments relate to a system that reduces bandwidth requirements for transmitting telemetry data from sensors in a computer system. During operation, the system obtains a cross-imputability value for each sensor in a set of sensors that are monitoring the computer system, wherein a cross-imputability value for a sensor indicates how well a sensor value obtained from the sensor can be predicted based on sensor values obtained from other sensors in the set. Next, the system clusters sensors in the set of sensors into two or more groups based on the determined cross-imputability values. Then, while transmitting sensor values from the set of sensors, for a group of sensors having cross-imputability values exceeding a threshold, the system selectively transmits sensor values from some but not all of the sensors in the group to reduce a number of sensor values transmitted.

Abstract: A method of processing data by one or more data processing systems for classification of the processed data into one or more predefined classifications, the method comprising: receiving by one or more data processing systems social profile data; binding by the one or more data processing systems based on the input social profile data, values of one or more attributes included in the social profile data to one or more parameters of a classifier executing on the one or more data processing systems; classifying data representing the user into one or more predefined classifications; for one of the predefined classifications into which the data representing the user is classified, identifying by the one or more data processing systems a candidate action included in the predefined classification and unassociated with the user; and transmitting an alert to notify the user of the candidate action.