Abstract: Apparatus and methods for processing inputs by one or more neurons of a network. The neuron(s) may generate spikes based on receipt of multiple inputs. Latency of spike generation may be determined based on an input magnitude. Inputs may be scaled using for example a non-linear concave transform. Scaling may increase neuron sensitivity to lower magnitude inputs, thereby improving latency encoding of small amplitude inputs. The transformation function may be configured compatible with existing non-scaling neuron processes and used as a plug-in to existing neuron models. Use of input scaling may allow for an improved network operation and reduce task simulation time.

Abstract: A collection of data that is extremely large can be difficult to search and/or analyze. Relevance may be dramatically improved by automatically classifying queries and web pages in useful categories, and using these classification scores as relevance features. A thorough approach may require building a large number of classifiers, corresponding to the various types of information, activities, and products. Creation of classifiers and schematizers is provided on large data sets. Exercising the classifiers and schematizers on hundreds of millions of items may expose value that is inherent to the data by adding usable meta-data. Some aspects include active labeling exploration, automatic regularization and cold start, scaling with the number of items and the number of classifiers, active featuring, and segmentation and schematization.

Abstract: Methods, computing systems and computer program products implement embodiments of the present invention that include selecting a training dataset including training instances having respective training features, and applying a classifier to the training dataset, thereby generating a training classification that assigns, to each of the training instances, one of a plurality of categories, the classifier having an expected classification. A classification bias is detected in the training classification relative to the expected classification, and in response to the classification bias, a calibration matrix is defined based on the training features, and the classification bias. A production dataset including production instances is selected, and the classifier and the calibration matrix are applied to the production dataset, thereby generating a production classification quantification that assigns, to each of the production instances, one of the plurality of categories.

Abstract: Described is a system and method for determining a classification of an application that includes initiating a stress test on the application, the stress test including a predetermined number of stress events, wherein the stress events are based on a network impairment. A response by the application to each stress event is identified and the application is classified as a function of the response into one of a first classification and a second classification, the first classification indicative of a normal application and the second classification indicative of an undesired application. If, the application is in the second classification, a network response procedure is executed.

Abstract: A method and system for identification of a deviation from a periodic behavior pattern in a sequence of multimedia content segments are provided. The system comprises receiving the sequence of multimedia content segments; generating at least one signature for each multimedia content segment of the sequence of multimedia content segments; comparing at least two signatures generated for at least two consecutive multimedia content segments to detect a periodic behavior pattern; upon detecting the periodic behavior pattern, comparing at least one signature generated for at least a subsequently received multimedia content segment to at least one signature representing the detected multimedia content segment to identify a deviation from the periodic behavior pattern; and upon identifying the deviation from the periodic behavior pattern, generating a notification with respect to the at least one deviation.

Abstract: Methods and apparatuses for balancing computing workload via migrating computing tasks are disclosed. An artificial neural network (ANN) is trained based on the workload distribution over time for a host. The ANN predicts the workload for the host, and an indication may be sent to migrate at least one computing task away from the host. The indication is sent when the method is operating in a proactive mode and when the predicted workload is outside of a desired operating range. Some embodiments monitor the workload; and automatically switch the method to the proactive mode, when a difference between the monitored workload and the predicted workload is small. Other embodiments monitor the workload; and automatically switch the method to a reactive mode, when the monitored workload is outside of a failsafe operating range for the particular host.

Abstract: Adaptive controller apparatus of a plant may be implemented. The controller may comprise an encoder block and a control block. The encoder may utilize basis function kernel expansion technique to encode an arbitrary combination of inputs into spike output. The controller may comprise spiking neuron network operable according to reinforcement learning process. The network may receive the encoder output via a plurality of plastic connections. The process may be configured to adaptively modify connection weights in order to maximize process performance, associated with a target outcome. The relevant features of the input may be identified and used for enabling the controlled plant to achieve the target outcome.

Abstract: The subject disclosure is directed towards the use of Monte Carlo (MC) procedures for computing the value of information (VOI), including with long evidential sequences. An MC-VOI algorithm is used to output a decision as to balancing the value and costs of collecting information in advance of taking action by running prediction model-based simulations to determine execution paths through possible states, and processing the results of the simulations/paths taken into a final decision.

Abstract: Methods and systems for selecting a cohort group or a patient at risk from a population of patients with mild cognitive impairment. The methods include using a computer configured to perform the steps: receiving normalized learning data from a portion of the population of patients; tuning a set of decision trees on the normalized learning data; receiving patient data from one or more patients of the population, wherein the patient data is independent from the learning data; classifying the patient data with the tuned set of decision trees to obtain patient threshold values; and displaying the patient threshold values.

Abstract: Instant message systems are enhanced by establishing at least one handling rule that is evaluated in an instant messaging environment, each established handling rule defining a condition based upon at least one of identified content or identified context, and a corresponding event handling action to be performed within the instant message environment. Upon identifying an instant message conversation within the instant message environment between a user and a correspondent, each handling rule is evaluated and the event handling action of an associated handling rule is performed if it is determined that the condition of handling rule is satisfied. Further, an indication that the corresponding event action was performed is conveyed to a corresponding user.

Abstract: A system and method that facilitates authoring of a Bayesian Belief Networks by: accessing text content stored in a content storage device; identifying statements within said accessed text content indicating a dependence relation; extracting said statements indicating said dependence relation from said text content; and aggregating said extracted statements into a form suitable for representation as a BBN network structure. To identify statements indicating a dependence relation, the method includes identifying one or more lexical and semantic attributes of variables within a text unit indicating a conditional dependence relation between two or more variables. The method further processes the text content to extract probabilistic information and probability statements and aggregate the probability statements into a quantitative layer of the BBN structure.

Abstract: A rules engine can be initialized as part of an object grid, wherein the rules engine is executable code executing using computer resources. The rules engine updates a business mapper. The rules engine receives a business rule definition to form a business rule. The rules engine determines business terms are present in the business rule. The rules engine converts the business terms to technical terms based on the business mapper, responsive to a determination that business terms are present. The rules engine determines that the business rule comprises a temporal rule. The rules engine receives rule criteria, wherein the rule criteria is at least one selected from the group consisting of frequency, delay and timing dependency, responsive to the determination that the business rule is a temporal rule. The rules engine stores the business rule to the rules engine. The rules engine publishes the business rule to a publish-subscribe topic.

Abstract: A system and method for the automated generation of documents for a legal transaction over a network using probabilistic prediction of customary usage. The predictions are generated by user experience, expert rules and machine learned classifiers based on user input of transaction data. The classifiers are constructed and tested on a partitioned dataset consisting of transaction data and legal document and clause selections in previous transactions. In one embodiment, such dataset is collected in a document management system.

Abstract: Techniques for presenting insight into classification trees may include performing a grouping analysis to group leaf nodes of a classification tree into a significant group and an insignificant group, performing influential target category analysis to identify one or more influential target categories for the leaf nodes of the classification tree in the significant group, and presenting one or more insights into the classification tree based on the grouping analysis and the influential target category analysis. Techniques for presenting insight into regression trees may include performing a grouping analysis to group leaf nodes of a regression tree into a high group and a low group, performing unusual node detection analysis to detect one or more outlier nodes in the high group and in the low group, and presenting one or more insights into the regression tree based on the grouping analysis and the unusual node detection analysis.

Abstract: The present invention relates to inspection of medical patients including, but not limited to, phonocardiography, auscultation and ultrasound medical imaging and other non-acoustical inspection techniques; and industrial non-destructive testing and evaluation of materials, structural components and machinery; and more particularly to the incorporation of cognitive artificial intelligence into an inspection system and method that utilizes cognitive mathematical techniques which emulate the cognitive processing abilities of the human brain including, but not limited to, symbolic cognitive architectures and inference process algebras, to analyze data collected from infrasound acoustical sensors (0.

Abstract: Embodiments relate to systems and methods for training a self-learning network using interpolated input sets based on a target output. A database management system can store sets of operational data, such as financial, medical, climate or other information. A user can input or access a set of target data, representing an output which a user wishes to be generated from an interpolated set of input data. The interpolation engine can generate a conformal interpolation function and input sets that map to the set of target output data. After interpolation, the interpolation engine can transmit the interpolated inputs, along with the set of target output data and other information, to a self-learning network such as a neural or fuzzy logic network. The self-learning network can be trained to converge to the target output based on the interpolated input results as generated by the interpolation engine, thus reproducing the desired interpolation function.

Abstract: A digital instructional environment leverages an infrared eye-tracker to monitor a learner's reading and viewing of text and simulations for subject matter. The system detects out-of-order reading/viewing patterns that could lead to poor comprehension. The digital learning environment communicates with other tutorial components including simulation environments, pedagogical agents and may respond in real-time to such patterns with messages that guide learners (knowledge acquirers) to return to effective reading/viewing patterns so as to promote effective construction of mental model(s) developed during knowledge acquisition/learning.

Abstract: Techniques for presenting insight into classification trees may include performing a grouping analysis to group leaf nodes of a classification tree into a significant group and an insignificant group, performing influential target category analysis to identify one or more influential target categories for the leaf nodes of the classification tree in the significant group, and presenting one or more insights into the classification tree based on the grouping analysis and the influential target category analysis. Techniques for presenting insight into regression trees may include performing a grouping analysis to group leaf nodes of a regression tree into a high group and a low group, performing unusual node detection analysis to detect one or more outlier nodes in the high group and in the low group, and presenting one or more insights into the regression tree based on the grouping analysis and the unusual node detection analysis.

Abstract: Predictive systems for designing enterprise applications include memory structures that output predictions to a user. The predictive system may include an HTM structure that comprises a tree-shaped hierarchy of memory nodes, wherein each memory node has a learning and memory function, and is hierarchical in space and time that allows them to efficiently model the structure of the world. The memory nodes learn causes, predicts with probability values, and form beliefs based on the input data, where the learning algorithm stores likely sequence of patterns in the nodes. By combining memory of likely sequences with current input data, the nodes may predict the next event. The predictive system may employ an HHMM structure comprising states, wherein each state is itself an HHMM. The states of the HHMM generate sequences of observation symbols for making predictions.

Abstract: Sensory encoder may be implemented. Visual encoder apparatus may comprise spiking neuron network configured to receive photodetector input. Excitability of neurons may be adjusted and output spike may be generated based on the input. When neurons generate spiking response, spiking threshold may be dynamically adapted to produce desired output rate. The encoder may dynamically adapt its input range to match statistics of the input and to produce output spikes at an appropriate rate and/or latency. Adaptive input range adjustment and/or spiking threshold adjustment collaborate to enable recognition of features in sensory input of varying dynamic range.