Abstract: The method of present disclosure relates to neural chip and optimizing operation of a neural chip. The method includes sensing current values of physical parameters indicating an environment. Sensed current values are stored in a memory unit. The memory unit also stores previously sensed values of physical parameters. The current values and the previously sensed values are compared by the neural chip. Based on the comparison, one or more actions are applied using the previously sensed values, for completing the task, if the current values and the previously sensed values are matched. In case there is no matching, the neural chip uses the current valises for applying the one or more actions. The neural chip learns from applying of the actions and updates itself accordingly.

Abstract: A system includes a processor configured to wirelessly receive data indicating vehicle-feature usage for an individual vehicle. The processor is also configured to aggregate received data to form a feature-usage customer profile defining feature preferences. The processor is further configured to select vehicles associated with a customer-classification, including predefined feature-usage characteristics, the customer-classification determined based on a correlation between the predefined feature-usage characteristics and the aggregated data in the feature-usage profile defining feature preferences. The processor is also configured to compare the aggregated data to the selected vehicles to determine a vehicle having features preferred by a customer as indicated by the aggregated data in the feature-usage profile and recommend the determined vehicle to the customer.

Abstract: A method includes performing by a processor: estimating a total cathode space current for a thermionic vacuum tube having at least one grid and a plate, such that at least one amplification factor associated with the at least one grid is determined by a polynomial based on a variable that represents at plurality of voltages associated with the at least one grid and the plate, the variable being heuristically determine. Transitions between positive and negative grid operation may experience a step change in estimated current value caused by the inclusion or elimination of grid current. A part of the grid current may be added back into the plate current during transition. This small contribution to plate current may gradually diminish as tube operation moves farther away from the transition boundary.

Abstract: Systems and methods for mapping configuration items to business functions within a corporate infrastructure are disclosed. Discovery processes to automatically create and update service maps may introduce an artificial dependency between configuration items that is not necessary to the business function represented in the service map. These unnecessary dependencies may be considered “noise” and unnecessarily complicate the service map. Using machine learning techniques and procedures to identify short lived connections embodiments in accordance with this disclosure, dependency connections that may be considered noise may be detected and flagged. Once detected, these connections may be automatically removed from the service map to improve its accuracy and usefulness. Additionally, a user interface is provided that explains the “reason codes” for identification of noise connections.

Abstract: A method is proposed for training a multitask computer system, such as a multitask neural network system. The system comprises a set of trainable workers and a shared module. The trainable workers and shared module are trained on a plurality of different tasks, such that each worker learns to perform a corresponding one of the tasks according to a respective task policy, and said shared policy network learns a multitask policy which represents common behavior for the tasks. The coordinated training is performed by optimizing an objective function comprising, for each task: a reward term indicative of an expected reward earned by a worker in performing the corresponding task according to the task policy; and at least one entropy term which regularizes the distribution of the task policy towards the distribution of the multitask policy.

Abstract: This invention, which focuses on personality and aptitude matching by psychophysiologic response to stimuli, is referred to as Brain Matching. In general terms, this invention starts by selecting highly specialized skill sets and top performer group for each skill set. The various groups are analyzed though psychophysiologic stimuli testing by using basically the same testing consisting of large numbers stimuli tests in a consistent testing environment. Stimuli tests can range from hundreds to thousands of images each producing a brainwave response. Neural Networks, Artificial Intelligence, Deep Learning computers look at the test results, highly specialized group by other highly specialized group to reduce the groups signature/response commonality into a template. Test subjects are then tested using the same stimuli. The subject's test results are analyzed for correlation with the various specialized expert groups.

Abstract: A DC voltage switch may include: a first node and a second node for series integration into a pole of a DC voltage line; a third node for the other pole of the line; a mechanical switch between the first and second nodes; a pulse-current module in parallel with the switch; four semiconductor switches connected as bridges comprising two series of two semiconductor switches; a pulse-current capacitor in parallel with the two series; and a switchable semiconductor element. The pulse-current module includes three module nodes. Potential points between the semiconductor switches of the two series correspond to the first and second module node and the outer ends of the two series of in each case two of the semiconductor switches are connected in pairs to a fourth module node and a fifth module node and the semiconductor element is between the fifth and third module node.

Abstract: A technique for training a neural network including an input layer, one or more hidden layers and an output layer, in which the trained neural network can be used to perform a task such as speech recognition. In the technique, a base of the neural network having at least a pre-trained hidden layer is prepared. A parameter set associated with one pre-trained hidden layer in the neural network is decomposed into a plurality of new parameter sets. The number of hidden layers in the neural network is increased by using the plurality of the new parameter sets. Pre-training for the neural network is performed.

Abstract: Systems and methods for probability forecasts and an energy transmission and/or energy distribution network are provided. Operational management may be carried out using a network control system with systematic consideration of forecast uncertainties. The probability of a distribution network being operable in a stable manner (e.g., with N-1 certainty) in a planning period is included. The system includes a forecaster for forecasts for a planning period, a forecast analyzer connected to the forecasts from the at least one forecaster, and elements for further information for outputting estimated forecast uncertainties. The system also includes a stability probability analyzer connected to the forecasts from the at least one forecaster, the estimated forecast uncertainties from the forecast analyzer, and elements for further information for outputting at least one item of information relating to an N-1 stability of the distribution network in the planning period.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for an crowd sourced training of an artificial intelligence system. One of the methods includes generating a training set using the customer communication information. The method includes training an artificial intelligence system using the training set. The method includes extracting at least one conversation pattern using the artificial intelligence system. The method includes the actions of instructing a chat application to process the at least one conversation pattern.

Abstract: Various embodiments described herein provide for a neural network tailored, based on user-provided input data, to detect user-specified objects in image data. An architecture of an embodiment may use unlabeled data from the user, such as a set of images from a video camera stream, while parameters of a tailored neural network (CNN) are trained or adapted.

Abstract: A system combines inputs from human processing and machine processing, and employs machine learning to improve processing of individual tasks based on comparison of human processing results. Once performance of a particular task by machine processing reaches a threshold, the level of human processing used on that task is reduced.

Abstract: A method and apparatus for generating information. A specific embodiment of the method includes: acquiring a set of geographic information point sequences and a set of identifiers comprising an identifier of each geographic information point sequence in the set of the geographic information point sequences; for the each geographic information point sequence in the set of the geographic information point sequences, clustering geographic information points in the each geographic information point sequence and generating an element sequence corresponding to the each geographic information point sequence; learning, by utilizing a machine learning method based on the set of identifiers and the generated element sequences, to obtain a matrix for the identifiers in the set of identifiers; and generating, for each identifier in the set of identifiers, based on the identifier and the matrix, information of a user to which the geographic information point sequence indicated by the identifier belongs.

Abstract: One embodiment provides for a method of transmitting data between multiple compute nodes of a distributed compute system, the method comprising creating a global view of communication operations to be performed between the multiple compute nodes of the distributed compute system, the global view created using information specific to a machine learning model associated with the distributed compute system; using the global view to determine a communication cost of the communication operations; and automatically determining a number of network endpoints for use in transmitting the data between the multiple compute nodes of the distributed compute system.

Abstract: A method for determining an architecture for a task neural network configured to perform a particular machine learning task is described.

Abstract: A mechanism is described for facilitating memory handling and data management in machine learning at autonomous machines. A method of embodiments, as described herein, includes detecting multiple tables associated with multiple neural networks at multiple autonomous machines, where each of the multiple tables include an index. The method may further include combining the multiple tables and multiple indexes associated with the multiple tables into a single table and a single index, respectively, where the single table is communicated to the multiple autonomous machines to allow simultaneous processing of one or more portions of the single table using one or more memory devices and one or more processors of one or more of the multiple autonomous machines.

Abstract: In one or more embodiments, one or more methods, processes, and/or systems may receive data associated with effective completion of tasks by agents and determine a positive correlation within the data between first particular feature values of feature vectors associated with the tasks and second particular feature values of feature vectors associated with the agents. A first agent associated with a feature vector that matches, within a first threshold, the second particular feature values may be selected, and a probability that the first agent will effectively complete a first task based on a feature vector associated with the first task matching, within a second threshold, the first particular feature values may be determined.

Abstract: A computer-implement method and an apparatus are provided for neural network reinforcement learning. The method includes obtaining, by a processor, an action and observation sequence. The method further includes inputting, by the processor, each of a plurality of time frames of the action and observation sequence sequentially into a plurality of input nodes of a neural network. The method also includes updating, by the processor, a plurality of parameters of the neural network by using the neural network to approximate an action-value function of the action and observation sequence.

Abstract: A facility for identifying combinations of feature and machine learning algorithm parameters, where each combination can be combined with one or more machine learning algorithms to train a model, is disclosed. The facility evaluates each genome based on the ability of a model trained using that genome and a machine learning algorithm to produce accurate results when applied to a validation data set by, for example, generating a fitness or validation score for the trained model and the corresponding genome used to train the model. Genomes that produce fitness scores that exceed a fitness threshold are selected for mutation, mutated, and the process is repeated. These trained models can then be applied to new data to generate predictions for the underlying subject matter.

Abstract: A neural network device includes a crossbar grid including first metal lines running in a first direction and second metal lines running transversely to the first metal lines and being electrically isolated from the first metal lines. An array of cross-over elements is included. Each cross-over element is connected between a first metal line and a second metal line. The cross-over elements each include a floating gate transistor device having a floating node. The floating node is configured to store a programmable weight value.