Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: A method for identifying defects in a multi-barrier wellbore includes receiving log data, the log data corresponding to one or more wellbore operations, the log data including data from at least one measurement modality corresponding to a present measurement modality. The method also includes training, using the log data, a machine learning model. The method further includes acquiring wellbore data, via the present measurement modality, during a logging operation. The method also includes processing at least a portion of the wellbore data using the trained machine learning model. The method includes identifying one or more features of interest in the wellbore data, via the trained machine learning model.

Abstract: A computer implemented method is described for conducting text sequence machine learning, the method comprising: receiving an input sequence x=[x1, x2, . . . , xn], to produce a feature vector for a series of hidden states hx=[h1, h2, . . . , hn], wherein the feature vector for the series of hidden states hx is generated by performing pooling over a temporal dimension of all hidden states output by the encoder machine learning data architecture; and extracting from the series of hidden states hx, a mean and a variance parameter, and to encapsulate the mean and the variance parameter as an approximate posterior data structure.

Abstract: A method includes obtaining training data for a classifier, the training data comprises one or more target classes, obtaining candidate background classes, selecting negative classes from the candidate background classes, wherein the negative classes exclude candidate background classes that are close to the target classes, wherein the negative classes exclude candidate background classes that are very different from the target classes, and wherein the negative classes include candidate background classes that are similar to the target classes, and training the classifier on a combined set of the selected negative classes and target classes.

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: A multi-dimensional task facial beauty prediction method and system, and a storage medium are disclosed. The method includes the steps of: at a training phase, using first facial images to optimize a shared feature extraction network for extracting shared features and to train a plurality of sub-task networks for performing facial beauty classification tasks; at a testing phase, extracting shared features of second facial images; inputting the shared features to the trained plurality of sub-task networks; and obtaining a first beauty prediction result based on first output results of the plurality of sub-task networks.

Abstract: A processor-implemented neural network method includes: generating a bit vector based on whether each of a plurality of input activations within a neural network is 0; merging the bit vector into the input activations such that bit values within the neural network included in the bit vector are most significant bits (MSBs) of multi bit expressions of the input activations; merging the bit vector into weights such that the bit values included in the bit vector are MSBs of multi bit expressions of the weights; sorting the input activations and the weights based on bits corresponding to the MSBs; and implementing the neural network, including performing operations between the sorted input activations and the sorted weights.

Abstract: An Industrial Virtual Assistant (IVA) platform with Robotic Process Automation that operates like a Digital Knowledge Companion and allows operational staff at industrial facilities to have natural language conversations with the IVA to obtain information about, and to control operations of, industrial facilities, and which automates certain processes based in part on those natural language conversations. In an embodiment, the platform uses a Robotic Process Automater (RPA) to ingest information from documentation, human inputs, and operational data from the facility, organize that information into a knowledge graph containing comprehensive facility information, and apply machine learning algorithms to the knowledge graph to provide natural language responses to human queries and to automate certain processes of the facility.

Abstract: A processor-implemented neural network method includes: generating a bit vector based on whether each of a plurality of input activations within a neural network is 0; merging the bit vector into the input activations such that bit values within the neural network included in the bit vector are most significant bits (MSBs) of multi bit expressions of the input activations; merging the bit vector into weights such that the bit values included in the bit vector are MSBs of multi bit expressions of the weights; sorting the input activations and the weights based on bits corresponding to the MSBs; and implementing the neural network, including performing operations between the sorted input activations and the sorted weights.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: Computer-implemented systems and methods for reducing an amount of computational resources consumed by a machine-learning model are provided. A machine-learning model is applied to a dataset to generate a first output. The machine-learning model includes a plurality of variables. Variables are iteratively removed from the machine-learning model, and for each iteration, the machine-learning model is applied with one or more variables removed from the dataset to generate a second output. For each iteration, the first and second outputs are compared. A subset of the removed variables having impact below a predetermined threshold on an output of the machine-learning model is determined based on the comparisons. An optimized machine-learning model that omits the subset of variables is applied to new data to generate an output for the new data.

Abstract: This deterioration estimation device is provided with: an SOH acquisition unit which acquires the SOH of a power storage element at a first time and the SOH at a second time after the first time; and a learning processing unit which trains a learning model on the basis of learning data which, as input data, includes time series data relating to the state of the power storage element from the first time to the second time, and the SOH at the first time and, as output data, includes the SOH at the second time.

Abstract: A method is provided that may include generating, based on a rule, an abstract syntax tree structure including a plurality of nodes, each of the plurality of nodes being associated with a portion of the rule; traversing a subset of the plurality of nodes and executing logic associated with each node of the subset; generating a plurality of execution events during the execution of the logic associated with each node of the subset; receiving a request to replay at least a part of the plurality of execution events; generating a user interface view to enable presentation of the rule as a plurality of nodes; and stepping through the plurality of execution events by selecting an execution event from the plurality of execution events and updating a presentation of a respective node identified by the selected execution event. Related systems, methods, and articles of manufacture are also disclosed.

Abstract: Systems and methods for accurately determining the state of health (including state of charge and relative age) of a Lithium Sulfur battery, module or cell. The invention uses an operational model of a Lithium Sulfur cell or battery to predict model parameters under a range of conditions related to state of charge and state of health. Operational models include the memory effect due to the unique chemistry of a Lithium Sulfur cell that precludes the user of other methodologies for State of health determination for Lithium Sulfur batteries. Model parameters are identified in real life applications and parameters are compared to those of the operational Lithium Sulfur model employing Kalman filtering. The output includes an estimate of state of health and other key performance indicators. Key performance indicators are compared with measured values of for example resistance to provide feedback to the estimate process in order to improve accuracy.

Abstract: There is provided a method for providing pricing information about a potential migration of a network computing task. The method includes the step of providing a provider availability set, identifying a price qualified set for each cloud service provider, and displaying the information about each cloud service provider and its pricing via a user interface. The price qualified set includes each cloud service provider: (a) whose price for performing the task complies with a price acceptability criteria or (b) that has a respective price proposal associated therewith reflecting a price for engaging the cloud service provider to perform the task in lieu of the task being performed by the network.

Abstract: According to one embodiment, a learning apparatus includes a first rule generator, a feature value calculator, a related word extractor, a second rule generator, and a learning unit. The first rule generator generates a first rule to label the event candidate, the first rule including a keyword of the event candidate. The feature value calculator calculates feature values of other words included in the text other than the event candidate. The related word extractor extracts a related word relating to the keyword from the other words using the feature values. The second rule generator generates a second rule to label the event candidate, the second rule being different from the first rule and including the related word. The learning unit generates learning data associating the keyword, the related word, and labeled event candidate with each other.

Abstract: A system and method for controlling a nodal network. The method includes estimating an effect on the objective caused by the existence or non-existence of a direct connection between a pair of nodes and changing a structure of the nodal network based at least in part on the estimate of the effect. A nodal network includes a strict partially ordered set, a weighted directed acyclic graph, an artificial neural network, and/or a layered feed-forward neural network.

Abstract: Systems and methods for allocating computing resources for a multi-tiered application are disclosed. A computer-implemented method includes: determining, by a computing device, a topology of a multi-tiered application; determining, by the computing device, a modeled setting of a computing resource for the multi-tiered application based on the determined topology; determining, by the computing device, an actual usage of the computing resource by the multi-tiered application; and adjusting, by the computing device, an allocation of the computing resource to the multi-tiered application based on the actual usage and the modeled setting.

Abstract: Automation security in a networked-based industrial controller environment is implemented. Various components, systems and methodologies are provided to facilitate varying levels of automation security in accordance with security analysis tools, security validation tools and/or security learning systems. The security analysis tool receives abstract factory models or descriptions for input and generates an output that can include security guidelines, components, topologies, procedures, rules, policies, and the like for deployment in an automation security network. The validation tools are operative in the automation security network, wherein the tools perform security checking and/or auditing functions, for example, to determine if security components are in place and/or in suitable working order.

Abstract: A device in a data processing system for training machine learning models receives a transaction and forwards it to at least one of a plurality of integrated control action models that use outputs of one model as inputs to other models. The models are machine learning models jointly trained for taking each control action of a plurality of control actions on the transaction to maximize an objective function based on probabilities of the control actions matching corresponding target control actions. The machine learning models include a risk model that outputs risk prediction information for a first control action that indicates whether or not to initiate processing of the transaction. The device further receives the risk prediction information from the risk model, and executes at least the first control action based on the risk prediction information.

Abstract: A method and system for servicing application requests from mobile devices that includes receiving, from a client device, a first application request to test a software application, obtaining, based on the first application request, a first application container package that includes a first production version of the software application and a first context data. The method and system further includes generating, according to the first context data, a first interface of the first production version of the software application, transmitting the first interface to the client device, receiving, from the client device, a second application request to use the software application, and transmitting, in response to the second application request, the production version of the software application.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: Embodiments of the present invention relate to efficiently computing variable predictiveness such that an indication of variable predictiveness can be provided in real time. In this regard, aspects of the present invention enable a user (e.g., digital marketer) to input a query and, in response, receive an indication of variable predictiveness. To efficiently compute variable predictiveness in response to a submitted user query, mutual information is computed offline and, thereafter, used to generate, in real time, conditional mutual information of variables for a specified date range. The concept of conditional mutual information can be utilized to represent variable predictiveness or otherwise indication variable predictiveness, such as to identify a set of variables that accurately predict a metric. Using such an approach effectively reduces a number of data access attempts and calculations performed in real time thereby reducing utilization of a processor(s).

Abstract: Systems and methods for allocating computing resources for a multi-tiered application are disclosed. A computer-implemented method includes: determining, by a computing device, a topology of a multi-tiered application; determining, by the computing device, a modeled setting of a computing resource for the multi-tiered application based on the determined topology; determining, by the computing device, an actual usage of the computing resource by the multi-tiered application; and adjusting, by the computing device, an allocation of the computing resource to the multi-tiered application based on the actual usage and the modeled setting.

Abstract: Mechanisms are provided for detecting interesting decision rules from a set of decision rules in a tree ensemble. Each tree in the tree ensemble is traversed in order to assign each individual data record from a set of data records to an identified leaf node in each tree. Predicted values are determined for the tree ensemble based on predictions provided by each leaf node to which each individual data record is assigned. Interesting sub-indices for decision rules from the set of decision rules are determined and, for each decision rule corresponding to the leaf nodes in the tree ensemble, the sub-indices are combined into interestingness index It. The decision rules are ranked corresponding to the leaf nodes in the tree ensemble according to the associated value of the interestingness index It and a subset of the decision rules corresponding to the leaf nodes in the tree ensemble are reported.

Abstract: Mechanisms are provided for detecting interesting decision rules from a set of decision rules in a tree ensemble. Each tree in the tree ensemble is traversed in order to assign each individual data record from a set of data records to an identified leaf node in each tree. Predicted values are determined for the tree ensemble based on predictions provided by each leaf node to which each individual data record is assigned. Interesting sub-indices for decision rules from the set of decision rules are determined and, for each decision rule corresponding to the leaf nodes in the tree ensemble, the sub-indices are combined into interestingness index It. The decision rules are ranked corresponding to the leaf nodes in the tree ensemble according to the associated value of the interestingness index It and a subset of the decision rules corresponding to the leaf nodes in the tree ensemble are reported.

Abstract: Systems and methods for generating data representative of insurance coverage for autonomous vehicles are provided. In particular, systems and methods for generating data representative of insurance coverage for autonomous vehicles based on data representative of autonomous vehicle reliability are provided. The data representative of autonomous vehicle reliability may include data representative of the autonomous vehicle original equipment manufacturer test results, autonomous vehicle system manufacturer test results, autonomous vehicle system component manufacturer test results, insurance company autonomous vehicle, systems and/or component test results, and/or third party test results of the autonomous vehicle, systems and/or components.

Abstract: Data matches are calculated between input data and repository data via a digest based matching algorithm where the reference digests corresponding to a repository interval of data identified as similar to an input interval of data are loaded into a sequential array and into a search structure. Each of the matching digests found using the search structure are extended using the sequential array of reference digests. Repository data intervals are determined as similar to an input data interval. Reference digests corresponding to the similar repository data interval are loaded into a sequential representation and into a search structure. Matches of input digests and the reference digests are found using the search structure. Each one of the found matches of the input digests and repository digests are extended using the sequential representation. Data matches are determined between the input data and the repository data using extended matches of digests.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a neural network used to select actions performed by a reinforcement learning agent interacting with an environment. In one aspect, a method includes maintaining a replay memory, where the replay memory stores pieces of experience data generated as a result of the reinforcement learning agent interacting with the environment. Each piece of experience data is associated with a respective expected learning progress measure that is a measure of an expected amount of progress made in the training of the neural network if the neural network is trained on the piece of experience data. The method further includes selecting a piece of experience data from the replay memory by prioritizing for selection pieces of experience data having relatively higher expected learning progress measures and training the neural network on the selected piece of experience data.

Abstract: An artificial hypothalamus system for regulating body temperature for a human can include monitoring data from a plurality of sensors. The sensor data can be used in a control system to optimize temperature regulation in real time through a feedback loop. The feedback loop can include monitoring the sensor data, evaluating a predetermined set of fuzzy rules using the data, and combining the output of the fuzzy rules to produce a precise value. The precise value can correspond to an output level for various temperature regulation devices, and a signal can be transmitted based on the precise value to activate the temperature regulation devices to a corresponding level of output.

Abstract: An apparatus for detecting code cloning of software includes a memory in which a program configured to detect code cloning from software is stored; and a processor configured to execute the program. Wherein the processor extracts multiple functions from a source code corresponding to the software, performs normalization and abstraction and compares the normalized and abstracted multiple functions with a vulnerable code clone set to determine whether there is code cloning of the software, and the vulnerable code clone set includes multiple vulnerable codes which are extracted from multiple functions included in other software, and normalized, abstracted and then stored.

Abstract: A computer performs a search. The computer receives a search request including a search term, and determines a plurality of association rules that are each a logical implication that the appearance of the search term follows from the appearance, in a corpus of data, of at least one keyword. The computer generates a search matrix having a plurality of cells, and having axes labeled with a plurality of keywords, wherein at least one cell of the search matrix is associated with an association rule of the plurality of association rules. Based on a cell selection of the search matrix, the computer generates an augmented search string including the search term and at least one keyword of the plurality of association rules.

Abstract: Robotic devices may be trained by a trainer guiding the robot along a target trajectory using physical contact with the robot. The robot may comprise an adaptive controller configured to generate control commands based on one or more of the trainer input, sensory input, and/or performance measure. The trainer may observe task execution by the robot. Responsive to observing a discrepancy between the target behavior and the actual behavior, the trainer may provide a teaching input via a haptic action. The robot may execute the action based on a combination of the internal control signal produced by a learning process of the robot and the training input. The robot may infer the teaching input based on a comparison of a predicted state and actual state of the robot. The robot's learning process may be adjusted in accordance with the teaching input so as to reduce the discrepancy during a subsequent trial.

Abstract: A face detector training method, a face detection method, and apparatuses are provided. In the present invention, during a training phase, a flexible block based local binary pattern feature and a corresponding second classifier are constructed, appropriate second classifiers are searched for to generate multiple first classifiers, and multiple layers of first classifiers that are obtained by using a cascading method form a final face detector; and during a detection phase, face detection is performed on a to-be-detected image by using a first classifier or a face detector that is learned during a training process, so that a face is differentiated from a non-face, and a face detection result is combined and output.

Abstract: Methods and systems for monitoring use, determining risk, and pricing insurance policies for an autonomous vehicle having one or more autonomous operation features are provided. According to certain aspects, accident risk factors may be determined for autonomous operation features of the vehicle using information regarding the autonomous operation features of the vehicle or other accident related factors associated with the vehicle. The accident risk factors may indicate the ability of the autonomous operation features to avoid accidents during operation, particularly without vehicle operator intervention. The accident risk levels determined for a vehicle may further be used to determine or adjust aspects of an insurance policy associated with the vehicle.

Abstract: An apparatus and method for managing health data through a user terminal are provided. The method includes inputting a user terminal number for identifying the user terminal, and information of a medical instrument for measuring the health data to a management server interworked with the user terminal, receiving, by a receiver, a security type table mapped onto the user terminal number and comprising a security type code for instructing the health data which the medical instrument has measured to be stored in a first memory without security or in a second memory with security, from the management server, and storing the health data in the first memory or the second memory, which the security type code instructs, through determining the security type code of the security type table when the health data is received from the medical instrument.

Abstract: Methods, systems, and techniques for managing groups of entities, such as individuals, employees, or systems, and providing entitlement and access to computer resources based on group membership are provided. Example embodiments provide a Group Management System having a Group Management Engine “GME,” an Entitlement Engine, and a Provisioning Engine, which work together to allow simplified grouping of entities and providing entitlement and access to the entities based upon the group membership. In one embodiment, the GME leverages dynamic programming techniques to enable accurate, scalable systems that can manage near real time updates and changes to the group's status or to the entities' status. These components cooperate to enable provisioning of applications based upon current entitlement.

Abstract: A method comprising using at least one hardware processor for constructing an index of corpora of multiple users, wherein each corpus of said corpora is associated with a single user of said multiple users, analyzing an electronic message addressed at least to one of said multiple users, to identify an expression which, based on the index, is likely to be unfamiliar or ambiguous to said one of the multiple users, and marking the expression in the electronic message.

Abstract: Systems and methods allow sharing and reuse of acquired customization knowledge of enterprise systems, utilizing collective customization experience. Particular embodiments employ Case-Based Reasoning (CBR) principles to manage customization experience in the context of both Software as a Service (SaaS) and Platform as a Service (PaaS) environments. Collective customization knowledge may be explicitly retrieved, adapted, shared, reused, and/or traded between tenants, partners, and optionally provider(s), of on-demand enterprise systems. A previous customization case stored within an experience base, may include one or more of the following: information allowing annotation of price information for trading the previous customization case (e.g.

Abstract: A method is provided for the open-loop or closed-loop control of at least one operating parameter of an electric energy accumulator influencing the aging state of the electric energy accumulator. The method determines the actual aging state of the electric energy accumulator, compares the actual aging state with a target aging state predefined for the momentary age of the energy accumulator, and restricts an operating parameter range permitted for the at least one operating parameter if the actual aging state is worse than the target aging state.

Abstract: A data classification system comprising a server adapted to receive data elements from a computer network, compute a distance between the elements and a plurality of representative elements such that where the distance is less than a threshold, the input elements are associated with the representative element and where the distance is not less than the threshold, the input element is stored as a new representative element. A method of classifying network traffic is further disclosed that can include providing a server adapted to receive data from a network, compute a distance between the data elements and representative data elements such that where the distance is less than a threshold, the input elements are associated with the representative element and where the distance is not less than the threshold, the input element is stored as a new representative element.

Abstract: The present technique relates to an information processor, an information processing method, and a program by which an objective variable value is efficiently and highly precisely estimated. A log acquisition unit acquires objective time-series data corresponding to the objective variable to be estimated, and a plurality of pieces of explanatory time-series data being time-series data corresponding to a plurality of explanatory variables explaining the objective variable. A model parameter update unit learns a parameter of a probability model using the acquired objective time-series data and plurality of pieces of explanatory time-series data. A log selection unit selects, based on the parameter of the probability model having been obtained by the learning, the explanatory variable corresponding to the explanatory time-series data acquired by the log acquisition unit.

Abstract: In one embodiment of the present invention, an apparatus may be used for supporting a survey based on information in an Internet medium. The apparatus comprises: a first acquisition hardware unit, wherein the first acquisition hardware unit acquires first evaluation information representing a degree of evaluation acquired by a survey of a real society pertaining to a prescribed target; a second acquisition hardware unit, wherein the second acquisition hardware unit acquires second evaluation information representing a degree of evaluation in the Internet medium pertaining to the prescribed target; and an estimator hardware unit, wherein the estimator hardware unit estimates a bias in information in the Internet medium based on a deviation of the second evaluation information from the first evaluation information.

Abstract: The present invention relates to a control system comprising a control interface between one or more wind turbine generators and a power grid, where the wind turbine generators are coupled to the power grid and contribute to the power production of the grid. The control interface is arranged to receive a set of event data. In embodiments, the set of event data may be any data available to a SCADA system. The set of event data is analyzed in terms of predetermined event rules comprising at least one predefined event condition and a set of adaptive event conditions. Based on the analysis an event output is provided in order to control a parameter of the one or more wind turbine generators. In embodiments, the control system may be implemented in, or in connection with a SCADA system, moreover, the event output may be based on fuzzy logic, a neural network or statistical analysis.

Abstract: A computerized method for optimizing parameters is described. A system can initialize a group of parameters to respective values within a set of allowable models and bound a partition function across a number of variable pairs to generate a plurality of bounds. The system can also determine new values for the group of parameters that minimize a sum of the plurality of bounds. The system can set the group of parameters to the new values and optimize the parameters by iteratively performing the bounding, determining and setting. The system can stop optimizing when a termination condition is reached.

Abstract: Technologies for redirecting traffic associated with a target entity for purposes of lawful intercept are presented herein. According to one aspect, a request to monitor data packets associated with a target entity for purposes of lawful intercept is received. Once the request is received, the target entity is added to a lawful intercept list and assigned a lawful intercept IP address from a range of lawful intercept IP addresses. Each lawful intercept IP address may have a corresponding routing policy for routing data packets associated with the corresponding lawful intercept IP address to a lawful intercept capture system. The data packet is then routed to the lawful intercept capture system based on the routing policy of the lawful intercept IP address. The lawful intercept capture system captures the data packet and forwards the data packet to a next hop network entity associated with the data packet.

Abstract: A bearing monitoring system includes at least one sensor, and a fuzzy logic enabled device operatively connected to the at least one sensor. The fuzzy logic enabled device includes a fuzzy logic rule set that is configured and disposed to determine a component condition based on an input from the at least one sensor.

Abstract: A system, method, and computer program product for combining replicated, analytical, and locally generated data in business objects on top of a database are disclosed. Business data are replicated from a first source to an analytics database. Analytical data based on the replicated business data are generated and combined with the replicated business data in the analytics database. Business objects are created to allow access to the analytics database via a user interface. Related apparatus, systems, techniques and articles are also described.

Abstract: Metrics representing a combined measure of power used by a central processing unit (CPU) and power used by a graphics processing unit (GPU) are compared to a shared supply power and thermal power budgets. Power used by the CPU and power used by the GPU are regulated in tandem using a fuzzy logic control system that can implement fuzzy rules that describe the management within thermal and supply power design constraints of the platform.

Abstract: Logic circuits provide networks to simulate the functions of neural networks of the brain, and can discriminate degrees of state, and combinations of degrees of state, corresponding to a number of neurons. Logic circuits comprise Recursive AND NOT Conjunctions (RANCs), or AND NOT gates. A RANC is a general logic circuit that performs conjunctions for 2n possible combinations of truth values of n propositions. The RANCs function dynamically, with capabilities of excitation and inhibition. Networks of RANCs are capable of subserving a variety of brain functions, including creative and analytical thought processes. A complete n-RANC produces all conjunctions corresponding to the 2n possible combinations of truth values of n propositions.