Abstract: Cooperative neural networks may be implemented by providing an input to a first neural network including a plurality of first parameters, and updating at least one first parameter based on an output from a recurrent neural network provided with the input, the recurrent neural network including a plurality of second parameters.

Abstract: Estimating Remaining Useful Life (RUL) from multi-sensor time series data is difficult through manual inspection. Current machine learning and data analytics methods, for RUL estimation require large number of failed instances for training, which are rarely available in practice, and these methods cannot use information from currently operational censored instances since their failure time is unknown. Embodiments of the present disclosure provide systems and methods for estimating RUL using time series data by implementing an LSTM-RNN based ordinal regression technique, wherein during training RUL value of failed instance(s) is encoded into a vector which is given as a target to the model. Unlike a failed instance, the exact RUL for a censored instance is unknown. For using the censored instances, target vectors are generated and the objective function is modified for training wherein the trained LSTM-RNN based ordinal regression is applied on an input test time series for RUL estimation.

Abstract: Embodiments employ an inference method for neural networks that enforces deterministic constraints on outputs without performing post-processing or expensive discrete search over the feasible space. Instead, for each input, the continuous weights are nudged until the network's unconstrained inference procedure generates an output that satisfies the constraints. This is achieved by expressing the hard constraints as an optimization problem over the continuous weights and employing backpropagation to change the weights of the network. Embodiments optimize over the energy of the violating outputs; since the weights directly determine the output through the energy, embodiments are able to manipulate the unconstrained inference procedure to produce outputs that conform to global constraints.

Abstract: Hardware and neural architecture co-search may be performed by operations including obtaining a specification of a function and a plurality of hardware design parameters. The hardware design parameters include a memory capacity, a number of computational resources, a communication bandwidth, and a template configuration for performing neural architecture inference. The operations further include determining, for each neural architecture among a plurality of neural architectures, an overall latency of performance of inference of the neural architecture by an accelerator within the hardware design parameters. Each neural architecture having been trained to perform the function with an accuracy. The operations further include selecting, from among the plurality of neural architectures, a neural architecture based on the overall latency and the accuracy.

Abstract: An embodiment of the invention provides a method to determine relationships between entities where an interface receives a first data set representing observations of a first entity and observations of a second entity, and a second data set representing a relationship between the first entity and the second entity. An entity analytics engine applies a first candidate rule to the first data set to generate a first candidate relationship between the first entity and the second entity. A processor determines whether according to a criterion the first candidate relationship matches the relationship represented in the second data set. The entity analytics engine replaces the first candidate rule with a second candidate rule by when the first candidate relationship does not match the relationship represented in the second data set.

Abstract: A method is provided, including: processing interactions by a plurality of users with a plurality of content items, the content items being provided over a network in response to user requests received over the network, wherein each content item is associated with one or more entities; for each user, determining a user entity set that includes entities associated with content items with which the user interacted; embedding the users and the entities in a vector space, wherein the embedding is configured to place a given user, and the entities of the given user's user entity set, in proximity to each other in the vector space; for each user, performing a proximity search in the vector space to identify a set of nearest entities to the user in the vector space; for each user, generating a user profile using the identified set of nearest entities to the user.

Abstract: Provided are a computer program product, a learning apparatus and a learning method. The method includes calculating a first propagation value that is propagated from a propagation source node to a propagation destination node in a neural network including nodes, based on node values of the propagation source node at time points and a weight corresponding to passage of time points based on a first attenuation coefficient. The method also includes updating the first attenuation coefficient by using a first update parameter, that is based on a first propagation value, and an error of the node value of the propagation destination node.

Abstract: A method for training a computerized mechanical device, comprising: receiving data documenting actions of an actuator performing a task in a plurality of iterations; calculating using the data a neural network dataset and used for performing the task; gathering in a plurality of reward iterations a plurality of scores given by an instructor to a plurality of states, each comprising at least one sensor value, while a robotic actuator performs the task according to the neural network; calculating using the plurality of scores a reward dataset used for computing a reward function; updating at least some of the neural network's plurality of parameters by receiving in each of a plurality of policy iterations a reward value computed by applying the reward function to another state comprising at least one sensor value, while the robotic actuator performs the task according to the neural network; and outputting the updated neural network.

Abstract: A method of processing image data in a connectionist network comprises a plurality of units, wherein the method implements a multi-channel unit forming a respective one of the plurality of units, and wherein the method comprises: receiving, at the data input, a plurality of input picture elements representing an image acquired by means of a multi-channel image sensor, wherein the plurality of input picture elements comprise a first and at least a second portion of input picture elements, wherein the first portion of input picture elements represents a first channel of the image sensor and the second portion of input picture elements represents a second channel of the image sensor; processing of the first and at least second portion of input picture elements separately from each other; and outputting, at the data output, the processed first and second portions of input picture elements.

Abstract: An apparatus includes processor(s) to: receive a request for a data catalog; in response to the request specifying a structural feature, analyze metadata of multiple data sets for an indication of including it, and to retrieve an indicated degree of certainty of detecting it for data sets including it; in response to the request specifying a contextual aspect, analyze context data of the multiple data sets for an indication of being subject to it, and to retrieve an indicated degree of certainty concerning it for data sets subject to it; selectively include each data set in the data catalog based on the request specifying a structural feature and/or a contextual aspect, and whether each data set meets what is specified; for each data set in the data catalog, generate a score indicative of the likelihood of meeting what is specified; and transmit the data catalog to the requesting device.

Abstract: An embodiment of the invention provides a method for determining relationships between physical entities, where one or more of the physical entities is associated with static feature(s) and changeable feature(s). An entity analytics engine determines that a first physical entity and a second physical entity may be in a relationship with a third physical entity based on a first rule and a first set of observations. The first rule is applicable to one or more static features of the first physical entity, the second physical entity, and the third physical entity. The first rule provides that the first physical entity and the second physical entity may be in a relationship with the third physical entity when the third physical entity includes one or more static features that are within a threshold degree of similarity to static features of the first physical entity and the second physical entity.

Abstract: Described herein is a technology that facilitates the production of and the use of automated datagens for event-based. A datagen (i.e., data-generator or data generation system) is a component, module, or subsystem of computer systems that searches, monitors, and analyzes machine data. A datagen produces events that are further processed in various ways for subsequent use (such as searching, monitoring, and analysis).

Abstract: A neuromorphic device includes a pre-synaptic neuron, a synapse electrically coupled to the pre-synaptic neuron through a row line, and a post-synaptic neuron electrically coupled to the synapse through a column line. The post-synaptic neuron includes an integrator, a comparator, and an error corrector including an error detector and a correction signal generator. The comparator and the error corrector receive an output of the integrator.

Abstract: Implementations of the present specification disclose methods, devices, and apparatuses for determining a feature interpretation of a predicted label value of a user generated by a GBDT model. In one aspect, the method includes separately obtaining, from each of a predetermined quantity of decision trees ranked among top decision trees, a leaf node and a score of the leaf node; determining a respective prediction path of each leaf node; obtaining, for each parent node on each prediction path, a split feature and a score of the parent node; determining, for each child node on each prediction path, a feature corresponding to the child node and a local increment of the feature on the child node; obtaining a collection of features respectively corresponding to the child nodes; and obtaining a respective measure of relevance between the feature corresponding to the at least one child node and the predicted label value.

Abstract: Disclosed aspects relate to entity model establishment using an infinite mixture topic modeling (IMTM) technique. A set of event data which corresponds to a set of events may be detected. Using the IMTM technique, the set of event data which corresponds to the set of events may be analyzed. Based on analyzing the set of event data using the IMTM technique, a set of entity models for the set of events may be determined. Based on the set of entity models for the set of events, a subset of the set of entity models for the set of events may be established.

Abstract: Systems and methods for quantifying temporal indeterminacy of timelines are provided. Systems and methods can rely on solving temporal constraint problems to extract timelines and can calculate the temporal relation loss during timeline transformation and then identify the temporal indeterminate sections of extracted timelines from both timelines and temporal graphs to measure the total temporal information loss.

Abstract: A configurable spiking neural network based accelerator system is provided. The accelerator system may be executed on an expansion card which may be a printed circuit board. The system includes one or more application specific integrated circuits comprising at least one spiking neural processing unit and a programmable logic device mounted on the printed circuit board. The spiking neural processing unit includes digital neuron circuits and digital, dynamic synaptic circuits. The programmable logic device is compatible with a local system bus. The spiking neural processing units contain digital circuits comprises a Spiking Neural Network that handles all of the neural processing. The Spiking Neural Network requires no software programming, but can be configured to perform a specific task via the Signal Coupling device and software executing on the host computer.

Abstract: Graph partitioning and placement for multi-chip neurosynaptic networks. According to various embodiments, a neural network description is read. The neural network description describes a plurality of neurons. The plurality of neurons has a mapping from an input domain of the neural network. The plurality of neurons is labeled based on the mapping from the input domain. The plurality of neurons is grouped into a plurality of groups according to the labeling. Each of the plurality of groups is continuous within the input domain. Each of the plurality of groups is assigned to at least one neurosynaptic core.

Abstract: Embodiments of the present invention provide an artificial neural network system for improved machine learning, feature pattern extraction and output labeling. The system comprises a first spiking neural network and a second spiking neural network. The first spiking neural network is configured to spontaneously learn complex, temporally overlapping features arising in an input pattern stream. Competitive learning is implemented as Spike Timing Dependent Plasticity with lateral inhibition in the first spiking neural network. The second spiking neural network is connected with the first spiking neural network through dynamic synapses, and is trained to interpret and label the output data of the first spiking neural network. Additionally, the output of the second spiking neural network is transmitted to a computing device, such as a CPU for post processing.

Abstract: A machine learning computing system for predicting a probability of success of an identified computing device error condition may include at least a first data repository storing a plurality of historic data records corresponding to one or more computing device error conditions and a second data repository storing a plurality of solutions to each of the computing device error conditions stored in the first data repository. A server is configured to receive a computing device error message from at least one computing center device and analyze the computing device error message to identify an associated error condition category. The server identifies at least two solutions to an associated error condition and predict a probability of success for each of the at least two solutions. The server then initiates at least one solution that has a greatest probability of success and updates the second data repository.