Abstract: In one embodiment, a service receives software version data regarding versions of software executed by devices in a network. The service detects a version change in the version of software executed by one or more of the devices, based on the received software version data. The service makes a determination that a drop in data quality of input data for a machine learning model used to monitor the network is associated with the detected version change. The service reverts the one or more devices to a prior version of software, based on the determination that the drop in quality of the input data for the machine learning model used to monitor the network is associated with the detected version change.

Abstract: Systems, methods and apparatus of intelligent optimization of caching operations in a data storage device. For example, the data storage device can include: one or more storage media components; a controller configured to store data into and retrieve data from the one or more storage media components; a cache memory configured to cache data that is stored in the one or more storage media components; and an artificial neural network configured to receive, as input and as a function of time, operating parameters indicative a data access pattern. The artificial neural network generates, based on the input, a prediction to determine an optimized set of cache configuration parameters; and cache operations of the cache memory are configured according to the optimized set of cache configuration parameters determined based on the prediction.

Abstract: One disclosed method involves determining first data indicative of a context of one or more client devices in response to access of a first file by the one or more client devices, and determining, subsequent to the access of the first file by the one or more client devices, second data indicative of a current context of a first client device. Based at least in part on the first data and the second data being of a first context type, at least one command is generated that causes the first client device to present an indication that the first file is available for access.

Abstract: A method is described for a proxy to mitigate attacks from web application clients based on context of web application layer requests. The method includes receiving a plurality of web application layer requests from a web application layer client; aggregating a first set of requests from the plurality of web application layer requests, wherein the first set of requests are part of a first session; determining a profile based on the first set of requests, wherein the profile describes a baseline of expected behavior for a user of the web application layer client; and determining a first threat value associated with the first set of requests based on the first set of requests and the profile, wherein the first threat value describes the likelihood that the first set of requests are part of an attack on one or more web application servers.

Abstract: A method and an apparatus for determining a status of a network device are provided. A warning analysis device obtains a plurality of pieces of target key performance indicator KPI data of the network device within preset duration, processes the plurality of pieces of target KPI data to generate an element, forms the feature vector by using generated elements corresponding to the plurality of pieces of feature information, and analyzes the feature vector based on a preset warning analysis model, to determine the status of the network device. In this way, the status of the network device is determined by analyzing a plurality of pieces of target KPI data within a period of time, instead of by using only data at a moment. This improves the accuracy of determining the network device, so as to reduce an omission of a warning.

Abstract: An autoscale-type performance assurance system performs autoscaling to increase or reduce the number of VMs/containers V1 to V4 generated in a server and resources of V1 to V4. A compute includes a plurality of types of V1 to V4, a data collection unit that collects a resource allocation amount of V1 to V4, and a resource control unit that performs autoscaling to increase or reduce the amount of resources of V1 to V4 according to a resource control amount. A controller includes a dependency calculation unit that calculates, based on the collected resource allocation amount, a degree of dependency indicating whether the resource allocation amount is dependent on a performance related to V1 to V4 for providing a communication service quality, and an autoscaling determination unit that obtains a resource control amount for increasing or reducing only resources related to the calculated degree of dependency indicating being dependent.

Abstract: In an example embodiment, a debugging tool is provided that includes logging functionality to allow a machine learned model administrator to replay machine learned recommendation model executions in order to identify points of error, without the scaling difficulties that would be involved in logging all features used in every recommendation.

Abstract: Some embodiments provide a non-transitory machine-readable medium that stores a program. The program receives a plurality of sets of words. Each set of words in the plurality of sets of words includes a word annotated as being an entity having a same custom entity type. The program further determines a plurality of word embeddings in a word embedding space for the plurality of annotated words. The program also defines a region in the word embedding space based on the received plurality of word embeddings. The program further receives a set of words. The program also determines a word embedding for a subset of the set of words. The program further determines whether the word embedding falls within the defined region in the word embedding space. Upon determining that the word embedding falls within the defined region in the word embedding space, the program also determines that the subset of the set of words represents an entity having the custom entity type.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating explainability data that explains a medical condition in a subject. In one aspect, a method comprises: obtaining data identifying a plurality of brain parcels that are predicted to be relevant to the medical condition; receiving fMRI data for a brain of a subject; processing the fMRI data for the brain of the subject to determine a respective activation score for each of the plurality of brain parcels that are predicted to be relevant to the medical condition; determining, for each of the plurality of brain parcels that are predicted to be relevant to the medical condition, a relative activation score for the brain parcel; and taking an action based on the relative activation scores.

Abstract: Methods and apparatus to facilitate generation of database queries are disclosed. An example apparatus includes a generator to generate a global importance tensor. The global importance tensor based on a knowledge graph representative of information stored in a database. The knowledge graph includes objects and connections between the objects. The global importance tensor includes importance values for different types of the connections between the objects. The example apparatus further includes an importance adaptation analyzer to generate a session importance tensor based on the global importance tensor and a user query, and a user interface to provide a suggested query to a user based on the session importance tensor.

Abstract: The invention discloses a MIMO different-factor full-form model-free control method. In view of the limitations of the existing MIMO full-form model-free control method with the same-factor structure, namely, at time k, different control inputs in the control input vector can only use the same values of penalty factor and step-size factors, the invention proposes a MIMO full-form model-free control method with the different-factor structure, namely, at time k, different control inputs in the control input vector can use different values of penalty factors and/or step-size factors, which can solve control problems of strongly nonlinear MIMO systems with different characteristics between control channels widely existing in complex plants. Compared with the existing control method, the inventive method has higher control accuracy, stronger stability and wider applicability.

Abstract: In a general aspect, information is encoded in data qubits in a three-dimensional device lattice. The data qubits reside in multiple layers of the three-dimensional device lattice, and each layer includes a respective two-dimensional device lattice. A three-dimensional color code is applied in the three-dimensional device lattice to detect errors in the data qubits residing in the multiple layers. A two-dimensional color code is applied in the two-dimensional device lattice in each respective layer to detect errors in one or more of the data qubits residing in the respective layer.

Abstract: Methods, systems and computer program products for improving ground truth quality for modeling are provided. Aspects include receiving a plurality of data inputs, wherein each of the plurality of data inputs has an associated label. Aspects also include training a model based on the plurality of data inputs. Aspects also include generating a plurality of vector representations corresponding to the plurality of data inputs based on the model. Aspects also include clustering the plurality of vector representations into one or more clusters. Aspects also include identifying at least one anomalous data input based on the one or more clusters. The at least one anomalous data input can be a data input of the plurality of data inputs that is mislabeled, contributes to an ambiguous class structure or is an outlier. Aspects also include outputting a notification that provides an indication of the at least one anomalous data input.

Abstract: Exemplary implementations may: generate output signals conveying contextual information and vehicle information; determine, based on the output signals, the contextual information; determine, based on the output signals, the vehicle information, determine, in an ongoing manner, based on the contextual information and/or the vehicle information, values of a complexity metric, the complexity metric quantifying predicted complexity of a current contextual environment and/or predicted complexity of a likely needed response to a change in the contextual information; filter, based on the values of the complexity metric, the contextual information spatially; and control, based on the vehicle information and the spatially filtered contextual information, the vehicle such that the likely needed response is satisfied.

Abstract: Techniques and solutions are described for determining relationships in data with improved efficiency, including computing resource use. A plurality of attributes are selected for analysis. The attributes can be processed, such as to facilitate relationship determination. Relationships between attribute values are determined. Redundant relationships can be removed. Distances are determined between relationships and used to select a sample of relationships. The sample is labelled by a user and used to train a machine learning classifier. The machine learning classifier labels determined relationships.

Abstract: Systems and methods for understanding (imputing meaning to) multi model data streams may be used in intelligent surveillance and allow a) real-time integration of streaming data from video, audio, infrared and other sensors; b) processing of the results of such integration to obtain understanding of the situation as it unfolds; c) assessing the level of threat inherent in the situation; and d) generating of warning advisories delivered to appropriate recipients as necessary for mitigating the threat. The system generates understanding of the system by creating and manipulating models of the situation as it unfolds. The creation and manipulation involve “neuronal packets” formed in mutually constraining associative networks of four basic types. The process is thermodynamically driven, striving to produce a minimal number of maximally stable models. Obtaining such models is experienced as grasping, or understanding the input steam (objects, their relations and the flow of changes).

Abstract: Logical rules can be automatically generated for use with event detection systems according to some aspects of the present disclosure. For example, a computing device can extract a group of logical rules from trained decision trees and apply a test data set to the group of logical rules to determine count values corresponding to the logical rules. The computing device can then determine performance metric values based on the count values, select a subset of logical rules from among the group of logical rules based on the performance metric values, and provide at least one logical rule in the subset for use with an event detection system. The event detection system can be configured to detect an event in relation to a target data set that was not used to train the decision trees.

Abstract: A low-latency database analysis system using an object index may include obtaining data expressing a usage intent, and, in response to a determination that the data expressing the usage intent includes object search request data including a user identifier and zero or more object search terms, obtaining, from an object-index responsive to the object search request data, object indexing data for an object, obtaining object detail data for the object, obtaining an object visualization capture for the object, and outputting object search response data including the object visualization capture and at least a portion of the object detail data for presentation to a user.

Abstract: A combination of two or more offers may be selected from a plurality of offers to be served to potential respondents, in a method performed using one or more processors in a computing system. The method of selecting may comprises receiving a request for a selection of a combination of two or more offers from the sets of offers; in response to the request: performing a series of selection operations each resulting in the selection of one offer to be included in the combination, each selection operation being carried out according to a respective model including one or more variables. The variables of the model for the second and any subsequent selection operation may include at least one selected offer from a previous selection operation. A signal identifying the selected offers may be output to cause the combination of the selected offers to be served to a respondent.

Abstract: A system includes a metric data store configured to receive and store a time-series of values of a first metric, a seasonal trend identification module configured to determine a periodicity profile for the first metric, and a modeling module configured to generate an autoregressive moving average (ARMA) model. The modeling module includes a seasonal model module configured to generate a first model of the time-series of values, a non-seasonal model module configured to generate a second model of the time-series of values, and a combination module configured to generate a third model based on the first and second models. The modeling module is configured to, in response to determining that a first periodicity profile describes the time-series of values, output the third model as the ARMA model. The system includes an envelope determination module configured to determine a normal behavior of the first metric based on the ARMA model.

Abstract: Methods and systems for self-certifying secure operation of a cyber-physical system having a plurality of monitoring nodes. In an embodiment, an artificial intelligence (AI) watchdog computer platform obtains, using the output of a local features extraction process of time series data of a plurality of monitoring nodes of a cyber-physical system and a global features extraction process, global features extraction data. The AI watchdog computer platform then obtains reduced dimensional data, generates an updated decision boundary, compares the updated decision boundary to a certification manifold, determines based on the comparison that the updated decision boundary is certified, and determines, based on an anomaly detection process, whether the cyber-physical system is behaving normally or abnormally.

Abstract: In an embodiment, an automated computer-based method for improving a computer system to be able to predict rejections of requisitions submitted to the computer system, the method comprising receiving a requisition from a client device; determining, at a scoring unit of a computer system, a probability value indicating a likelihood that the requisition would be rejected if the requisition is submitted to a requisition approval chain; transmitting the probability value from the computer system to the client device to be displayed on a display of the client device.

Abstract: According to one embodiment, a medical image processing apparatus of an embodiment includes an acquirer, a reliability setter and a learner. The acquirer acquires training data created by a creator on the basis of a medical image. The reliability setter sets, to the training data acquired by the acquirer, reliability information based on a creation situation of the training data or information about the creator who created the training data. The learner generates a learned model using the training data according to weighting based on the reliability information set by the reliability setter.

Abstract: Provided are systems, methods and computer program products for evaluating subsystem performance. In some embodiments, a method comprises perturbing a first attribute of a first subsystem of a system that includes a plurality of subsystems, determining a change in a second attribute of a second subsystem of the system in response to the perturbing of the first attribute, where at least one output of the first subsystem is passed to the second subsystem, and determining a value for a performance metric of the system based on a correlation of the performance metric with the first and second attributes. In some embodiments, the system is a software stack of an autonomous vehicle (AV) and the performance metric is an objective function output that measures a quality of the AV's driving behavior.

Abstract: In a robot, an actuator causes the robot to operate. A processor is configured to acquire, when a holding portion is held by a predetermined target, physical information on a physical function of the predetermined target, and cause, by controlling the actuator depending on the acquired physical information, the robot to perform at least one of an examination operation for examining the physical function of the predetermined target and a training support operation for training the physical function of the predetermined target.

Abstract: A method and apparatus for scoring member data in a social networking service is provided. A method comprises receiving input that indicates a particular feature that is not in a set of features upon which a first model is based. In response to receiving the input, training a plurality of models based on training data and a plurality of features that includes the set of features and the particular feature, and selecting, based on one or more criteria, a particular model from among the plurality of models that includes a second model that is of the same type as the first model. Using the particular model to score a particular data set based on the plurality of features indicated in the particular data set.

Abstract: The simplification of neural network models is described. For example, a method for simplifying a neural network model includes providing the neural network model to be simplified, defining a first temporal filter for the conveyance of input from a neuron to an other spatially-extended neuron along the arborized projection, defining a second temporal filter for the conveyance of input from yet another neuron to the spatially-extended neuron along the arborized projection, replacing, in the neural network model, the first, spatially-extended neuron with a first, spatially-constrained neuron and the arborized projection with a first connection extending between the first, spatially-constrained neuron and the second neuron, wherein the first connection filters input from the second neuron in accordance with the first temporal filter and a second connection extending between the first spatially-constrained neuron and the third neuron.

Abstract: System and methods for generating a trajectory of a dynamical system are described herein. An example method includes modelling a policy from demonstration data. The method also includes generating a first set of action particles by sampling from the policy, where each of the action particles in the first set includes a respective system action, predicting a respective outcome of the dynamical system in response to each of the action particles in the first set, and weighting each of the action particles in the first set according to a respective probability of achieving a desired outcome. The method further includes generating a second set of action particles by sampling from the weighted action particles, where each of the action particles in the second set includes a respective system action, and selecting a next system action in the trajectory of the dynamical system from the action particles in the second set.

Abstract: A platform identifies and matches software application programs for a business user based on his/her context such as industry, location, size, etc. by providing nuanced and personalized guidance for the business user to define the business problem and the capabilities needed to solve the problem. The platform permits use of custom dimensions to map software application programs to a business user's needs, fitment etc. to help him/her find a software program that matched to the business user's needs. The platform further harvest insights from a broad spectrum of inputs, such as thought leaders, to help a business user get insight on which software program best suits him/her. The platform provides a social platform for business users to collaborate and find the best software program for their needs.

Abstract: The present invention provides a method for learning latest data considering external influences in an early warning system, and the early warning system for same. The method for learning latest data considering external influences comprises the steps of: an early warning processing device categorizing device monitored variables according to external influences; and the early warning processing device differently applying a pattern learning method for each of the categorized monitored variables.

Abstract: A method is provided for managing a machine learning system. In the method, a database is provided for storing a plurality of data elements. A plurality of machine learning models is trained using assigned subsets of the plurality of data elements. The outputs of the plurality of machine learning models is provided to an aggregator. During inference operation of the machine learning system, the aggregator determines a final output based on outputs from the plurality of models. If it is determined that an assigned subset must be changed because, for example, a record must be deleted, then the data element is removed from the selected assigned subset. The affected machine learning model associated with the changed assigned subset is removed, and retrained using the changed assigned subset.

Abstract: An electronic apparatus and a compression method for an artificial neural network are provided. The compression method is adapted for the artificial neural network with a plurality of convolution layers. The compression method includes: setting a first pruning layer for coupling the first pruning layer to Lth convolution layer, where the first pruning layer has a plurality of first weighting values and each of the first weighting values corresponds to each of a plurality of channels of the Lth convolution layer; tuning the first weighting values, selecting a part of the channels of the Lth convolution layer to be at least one first redundancy channel according to the first weighting values, and generating a compressed Lth convolution layer by deleting the at least one first redundancy channel; and removing the first pruning layer, and generating a first compressed artificial neural network.

Abstract: Methods and apparatus for generating a facial feature. A specific embodiment of the method includes: acquiring a to-be-recognized face image; inputting the to-be-recognized face image into a first convolutional neural network to generate a feature area image set of the to-be-recognized face image, the first convolutional neural network being used to extract a feature area image from a face image; inputting each feature area image in the feature area image set into a corresponding second convolutional neural network to generate an area facial feature of the feature area image, the second convolutional neural network being used to extract the area facial feature of the corresponding feature area image; and generating a facial feature set of the to-be-recognized face image based on the area facial feature of the each feature area image in the feature area image set.

Abstract: An unsupervised method is presented for determining size of an object captured in an image. The method includes: capturing a set of images with a camera, where each image in the set of images is captured with the camera set at a different f-number; for each image in the set of images, computing a quality metric for the image, where the qualitymetric quantifies focus using contrast; identifying a given image in the set of images using the contrast metrics and the f-number of the camera; calculating a hyperfocal distance using parameters of the camera when taking the given image; and quantifying one or more dimensions of the object as a function of the hyperfocal distance; and if the object is known, using a view classifier to estimate the dimension in 3D.

Abstract: A device configured to emulate a quadrilateral lattice correlithm object generator includes multiple processing stages that operate together to output a quadrilateral lattice correlithm object. A quadrilateral lattice correlithm object has a quadrilateral geometric shape and is formed by at least four sub-lattice correlithm objects that are some number of bits away from each other in n-dimensional space.

Abstract: A computer-implementable method for managing a cognitive graph comprising: receiving data a data source, the data comprising a query and information relating to an answer to the query; processing the query, the processing the query identifying a plurality of query related knowledge elements; processing the information relating to the answer to the query, the processing the information relating to the answer to the query identifying a plurality of answer related knowledge elements; and, storing the plurality of query related knowledge elements and the plurality of answer related knowledge elements within the cognitive graph as a collection of knowledge elements.

Abstract: A data analysis device that analyzes data having a record including an objective variable and a plurality of explanatory variables includes a node generating unit that generates a node specified by a condition of the explanatory variable on the basis of the objective variable and the explanatory variable of the record and associating the record with the node, an evaluation value generating unit that generates a proportion of the number of records whose target value is the objective variable among a plurality of records associated with the node as an evaluation value, and a parameter extracting unit that selects a node on the basis of the evaluation value and extracts and outputs the condition of the explanatory variable related to the selected node.

Abstract: An apparatus includes a substrate, a classical computing processor formed on the substrate, a quantum computing processor formed on the substrate, and one or more coupling components between the classical computing processor and the quantum computing processor, the one or more coupling components being formed on the substrate and being configured to allow data exchange between the classical computing processor and the quantum computing processor.

Abstract: Techniques facilitating simplified quantum programming are provided. In one example, a computer-implemented method comprises reducing, by a device operatively coupled to a processor, a first computing problem of a problem type to a second computing problem of the problem type, wherein the second computing problem is associated with a quantum circuit; facilitating, by the device, execution of the quantum circuit at a quantum computer, resulting in a first output corresponding to the second computing problem; and mapping, by the device, the first output to a second output corresponding to the first computing problem.

Abstract: In one example, a processor receives a structured query, parses the structured query into components, and stores the structured query, the components, and at least one attribute regarding the structured query in a first query record of a query record storage platform, the at least one attribute comprising a user identification of a user generating the structured query. The processor may then receive a search associated with the query record storage platform, where the search includes at least one parameter, and where the at least one parameter specifies at least one of the structured query, at least one of the components, or at least one of the at least one attribute regarding the structured query, and return the first query record in response to the search.

Abstract: An aspect includes querying, by a processor, a plurality of model data from a distributed data source based at least in part on one or more user characteristics. A plurality of sensor data is gathered associated with a condition of a user. A policy is generated including an end goal and one or more sub-goals based at least in part on the model data and the sensor data. The policy is iteratively adapted based at least in part on one or more detected changes in the sensor data collected over a period of time to adjust at least one of the one or more sub-goals. The policy and the one or more sub-goals are provided to the user.

Abstract: A computer-implemented method for analyzing electronic documents includes defining at least one search criterion for a plurality of index items relating to a subject; applying the at least one search criterion to a plurality of documents; determining, for each index item in the plurality of index items, a category index reflecting a degree to which the plurality of documents relates to the index item; and generating, with reference to the category index for each index item in the plurality of index items, a performance index reflecting a degree to which the plurality of documents relates to the subject.

Abstract: A security system protecting a monitored system (e.g., a database server) is configured to self-update (extend) a statement/command parser grammar, dynamically. To accomplish this, the security system uses the monitored system itself as a syntax validator (on the parser's behalf), and without requiring any changes or modifications to the monitored server. In one embodiment, the security system comprises a protocol analyzer and the parser. The protocol analyzer extracts a statement/command from a received request and passes it to the parser. If no parser syntax error is found, the statement/command is validated against a security policy. If a parser syntax error occurs, however, the system examines a response from the monitored system to determine whether the parser syntax error is a “false positive.” If so, the parser then self-extends its own grammar automatically to correct for the parser syntax error. Once its grammar is updated, the parser is then able to process the original request.

Abstract: A method, computer-readable medium, and apparatus for classifying IP network traffic are disclosed. For example, a method may include a processor for applying a regularized probabilistic model to internet protocol network traffic for performing geolocation, wherein the internet protocol network traffic contains internet protocol address information, detecting an outlier in the internet protocol network traffic, wherein the outlier is associated with a transaction over a network, and performing a remedial action to address the outlier.

Abstract: A system and method are disclosed to facilitate automated database system workload classification. A utilization metrics data source may contain utilization metrics vectors associated with workloads of the database system. A variational autoencoder may receive utilization metrics vectors from the utilization metrics data source and encode the utilization metrics vectors into latent vector features. Moreover, the variational autoencoder may be trained to generate appropriate distributions around the latent vector features. A synthetic workload creation platform receives information about the distributions around the latent vector features and samples different values from the distributions to create synthetic workload vectors. A workload classification platform, trained using the synthetic workload vectors, may then generate workload classification labels for workloads running on the database system (e.g., to tune database parameters as appropriate).

Abstract: Provided is a sound event recognition method that may improve a sound event recognition performance using a correlation between difference sound signal feature parameters based on a neural network, in detail, that may extract a sound signal feature parameter from a sound signal including a sound event, and recognize the sound event included in the sound signal by applying a convolutional neural network (CNN) trained using the sound signal feature parameter.

Abstract: A method of detecting smoke or a fire involves the installation of properly produced cameras in the field. The cameras send recorded photos to a server. A computer downloads the photos from the server and uses attempts to automatically detect smoke with artificial intelligence software. Photos with a detected fire are sent to users which interface with a non-transitory computer readable medium, such as a desktop computer or a mobile phone, capable of executing a software application for verification. Verified smoke or fire threats signal alarms to the designated fire department in an effort to prevent fires from spreading and thus causing unnecessary damage to communities, lives, and ecosystems.

Abstract: A method for generating database queries from declarative queries having a known syntax. The method includes a query preparation software receiving a declarative query in a known system form. The query preparation software then analyzes the declarative query to build a set of generic query tasks with identified data types. The query preparation software then optimizes the set of generic query tasks and builds a target database query from the optimized generic query tasks reflecting features of a target database. The method further includes the query preparation software submitting the target database query to the target database.

Abstract: Evaluating risk of sensitive data associated with a target data set includes a computer system receiving a pattern that defines sensitive data and a selection of a data set as the target data set for evaluating. The system determines portions of the target data set from which to select sample data sets and determines, responsive to a confidence limit and sizes of the respective portions of the target data, a size of a sample data set for each respective target data set portion. The system randomly samples the target data set portions to provide sample data sets of the determined sample data set sizes and determines whether there is an occurrence of the sensitive data in each sample data set by searching for the pattern in the sample data sets. The system determines a proportion of the sample data sets that have the occurrence of the sensitive data.

Abstract: The present invention provides a method, computer program product, and system of generating prioritized list. In an embodiment, the method, computer program product, and system include receiving, by a computer system, target user identification data identifying a target user, target action data, social network content for the one or more users, and social network activity data for the one or more users, analyzing, by a computer system, social network links between the source user and the target user and the social network activity data for the one or more users, determining, by a computer system, a prioritized list of probabilistic action paths that could result in the target user performing the target action on the content based on the analyzing, and outputting the prioritized list to the source user.