Abstract: Systems and methods for non-invasive assessment of an arterial stenosis, comprising include segmenting a plurality of mesh candidates for an anatomical model of an artery including a stenosis region of a patient from medical imaging data. A hemodynamic index for the stenosis region is computed in each of the plurality of mesh candidates. It is determined whether a variation among values of the hemodynamic index for the stenosis region in each of the plurality of mesh candidates is significant with respect to a threshold associated with a clinical decision regarding the stenosis region.

Abstract: A highlight learning technique is provided to detect and identify highlights in sports videos. A set of event models are calculated from low-level frame information of the sports videos to identify recurring events within the videos. The event models are used to characterize videos by detecting events within the videos and using the detected events to generate an event vector. The event vector is used to train a classifier to identify the videos as highlight or non-highlight.

Abstract: An object detection system for an autonomous vehicle processes sensor data, including one or more images, obtained for a road segment on which the autonomous vehicle is being driven. The object detection system compares the images to three-dimensional (3D) environment data for the road segment to determine pixels in the images that correspond to objects not previously identified in the 3D environment data. The object detection system then analyzes the pixels to classify the objects not previously identified in the 3D environment data.

Abstract: There is provided a framework including systems and methods for analyzing skin parameters from images or videos showing skin. Using a series of Hierarchical Differential Image Filters (HDIF), it becomes possible to detect different skin features such as wrinkles, spots, and roughness. The hierarchical differential image filter computes two enhancements to an image showing skin at two different levels of enhancement, determines a differential image using the two enhancements and computes the skin analysis rating using the differential image. These skin ratings are comparably accurate to actual ratings by dermatologists.

Abstract: Methods and systems of an intelligent nanny assistant are described. A method may involve periodically or continuously receiving image-related data from a monitoring system that monitors an environment. The method may also involve determining a subject in the environment as a subject of concern and determining a range of sight of the subject of concern. The method may further involve retrieving information related to one or more objects of interest of the subject of concern. The method may involve controlling one or more devices in the environment to provide the information in a way that attracts the subject of concern to move away from a predefined area of the environment.

Abstract: Methods and systems of an intelligent nanny assistant are described. A processor may receive image-related data from a monitoring system and determine whether a subject of concern is approaching a predefined area of the environment based on the image-related data. In response to a determination that the subject of concern is approaching the predefined area, the processor may control an information output system to provide visual information, audible information, or both the visual information and the audible information in a way that attracts the subject of concern to move away from the predefined area.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classification using a neural network. One of the methods for processing an input through each of multiple layers of a neural network to generate an output, wherein each of the multiple layers of the neural network includes a respective multiple nodes includes for a particular layer of the multiple layers: receiving, by a classification system, an activation vector as input for the particular layer, selecting one or more nodes in the particular layer using the activation vector and a hash table that maps numeric values to nodes in the particular layer, and processing the activation vector using the selected nodes to generate an output for the particular layer.

Abstract: A classification system classifies different aspects of content of an input image stream, such as faces, landmarks, events, and so forth. The classification system includes a general classifier and at least one specialized classifier template. The general classifier is trained to classify a large number of different aspects of content, and a specialized classifier can be trained based on a specialized classifier template during operation of the classification system to classify a particular subset of the multiple different aspects of content. The classification system determines when to use the general classifier and when to use a specialized classifier based on class skew, which refers to the temporal locality of a subset of aspects of content in the image stream.

Abstract: An apparatus and method to facilitate providing recommendations are disclosed herein. A query image is received via a user interface. An image attribute from the query image is extracted. A determination that a usage condition for an image index is satisfied by the query image based on a comparison of an image attribute of the image index to the image attribute of the query image is performed. In response to the determination that the usage condition for the image index is satisfied by the query image, the image index is selected. Based on the selected image index and the usage condition, a set of item images from a plurality of item images stored in a database that correspond to the selected image index is identified. The identified set of item images are presented at the user interface.

Abstract: A system for labeling an object uses at least one object label made from a material that absorbs and reflects incident energy uniformly across all wavelengths of incident energy at a ratio proportional to a thickness of the material and that includes a pattern having variations in the thickness of the material along at least one of two orthogonal directions across the label. An interrogator directs a predetermined wavelength of radiation to the at least one label, and a reader to receives reflected radiation from the label at the predetermined wavelength and interprets the reflected radiation to recognize the pattern.

Abstract: A method for visual transmission and reception of repeatedly changing information comprises repeatedly performing the following set of actions, receiving a first set of information to transmit, the first set of information is an update of a previously received set of information, generating an information carrying first pattern representing the first set of information; displaying the information carrying first pattern in a first color; receiving a second set of information to transmit, the second set of information is an update of the first set of information; generating an information carrying second pattern representing the second set of information; displaying the information carrying second pattern in a second color, the second color differs from the first color; and wherein one of the information carrying patterns is statically displayed when another one of the information carrying patterns is updated.

Abstract: An electronic transaction card communicates with an add-on slot of an intelligent electronic device. The add-on slot may be a memory card slot. The intelligent electronic device may be a mobile phone or other device with or without network connectivity. The electronic transaction card may have magnetic field producing circuitry compatible with magnetic card readers, smartcard circuitry, other point-of-sale interfaces, or any combination thereof.

Abstract: A computer-implemented method includes identifying one or more tag readers. The one or more of tag readers are positioned peripherally to a region. The computer-implemented method further includes receiving a signal, by at least one of the one or more tag readers, from a tag. The computer-implemented method further includes determining a received signal strength indication for at least one of the one or more tag readers and mapping at least one received signal strength indication onto at a one dimensional representation of the region to yield a mapped data. The computer-implemented method further includes performing pattern recognition on the mapped data to yield a pattern data and applying machine learning to the pattern data to locate the tag within the region. A corresponding computer system and computer program product are also disclosed.

Abstract: Radio frequency identification (RFID) enabled devices are presented. In some examples, a RFID tag may be arranged on the device and the device may further include a RFID tag mode modifier which is configured to transfer the mode or state of the RFID tag from a first mode, such as an inactive mode, to a second mode, such as an active or enabled mode, or vice versa. For instance, the RFID tag mode modifier may transfer the RFID tag from an inactive mode or state to an active mode or state in which the RFID tag may be enabled for use, or vice versa. The RFID tag mode modifier may be an activation button, a contact region, or a physical device configured to prevent transmission of RFID signals. One or more methods of generating or using the devices described herein are also provided.

Abstract: A stored value card and carrier assembly including a carrier formed from one or more panels, at least one stored value card sandwiched and sealed within the carrier, and a label removably adhered or coupled to the card and adhered or coupled to an interior of the carrier such that the bond to the carrier or card is greater than the bond to the other of the card or carrier. The label includes at least a portion of activation indicia, such as a barcode, either printed or otherwise applied thereon. Upon removal of the card from the carrier, the label remains coupled to one of the carrier or the card, whichever bond is greater, rending any remaining activation indicia on the other of the stored value card or carrier unreadable.

Abstract: An electronic price indicating system according to the embodiment includes: at least one electronic price indicator for displaying product information; and a power supply apparatus connected to the at least one electronic price indicator to supply a driving signal which is driving power of the at least one electronic price indicator. The power supply apparatus includes: a power supply unit connected to the at least one electronic price indicator and supplying driving power to the electronic price indicator; an antenna for receiving data transmitted from a server; and a communication unit for confirming a first electronic price indicator to which data received via the antenna is to be applied and supplying power to the confirmed first electronic price indicator, wherein the communication unit controls an operation of the power supply unit, modulates the power supply signal according to the received data, and supplies the modulated power supply signal to the first electronic price indicator.

Abstract: A substrate for contactless transaction card, which is equipped with an integrated chip (IC) connected to at least one organic light-emitting diode (OLED) and an inductively powered tag.

Abstract: A technology for path determination using robust optimization is provided. In accordance with one aspect, a network graph of a network is generated. The network graph comprises nodes corresponding to points in the network, and edges which connect the nodes. Costs for each edge of the network are determined and modeled using reference point, upper bound and lower bound parameters. A user input which includes a source node, destination node, and cost target may be received from a client device. A resultant path connecting the source node and destination node are determined by solving a target-oriented robust optimization problem, which optimizes a cost of the resultant path based on the modeled costs of the edges. The resultant path is displayed on a user interface of the client device.

Abstract: A method and system are disclosed for improving a policy for a stochastic control problem, the stochastic control problem being characterized by a set of actions, a set of states, a reward structure as a function of states and actions, and a plurality of decision epochs, the method comprising using a sampling device obtaining data representative of sample configurations of a Boltzmann machine, obtaining initialization data and an initial policy for the stochastic control problem; assigning data representative of an initial weight and a bias of respectively each coupler and each node and the transverse field strength of the Boltzmann machine to the sampling device; until a stopping criterion is met generating a present-epoch state-action pair, amending data representative of none or at least one coupler and at least one bias, performing a sampling corresponding to the present-epoch state-action pair to obtain first sampling empirical means, obtaining an approximation of a value of a Q-function at the present-e

Abstract: A method in a hardware implementation of a Convolutional Neural Network (CNN), includes receiving a first subset of data having at least a portion of weight data and at least a portion of input data for a CNN layer and performing, using at least one convolution engine, a convolution of the first subset of data to generate a first partial result; receiving a second subset of data comprising at least a portion of weight data and at least a portion of input data for the CNN layer and performing, using the at least one convolution engine, a convolution of the second subset of data to generate a second partial result; and combining the first partial result and the second partial result to generate at least a portion of convolved data for a layer of the CNN.

Abstract: A method of configuring a hardware implementation of a Convolutional Neural Network (CNN), the method comprising: determining, for each of a plurality of layers of the CNN, a first number format for representing weight values in the layer based upon a distribution of weight values for the layer, the first number format comprising a first integer of a first predetermined bit-length and a first exponent value that is fixed for the layer; determining, for each of a plurality of layers of the CNN, a second number format for representing data values in the layer based upon a distribution of expected data values for the layer, the second number format comprising a second integer of a second predetermined bit-length and a second exponent value that is fixed for the layer; and storing the determined number formats for use in configuring the hardware implementation of a CNN.

Abstract: A method of generating mapping dictionaries for neural networks may be provided. A method may include receiving, at a current layer, encoded activation addresses from a previous layer and encoded weight addresses. The method may also include decoding the encoded activation addresses to generate decoded activation addresses, and decoding the encoded weight addresses to generate decoded weight addresses. Further, the method may include generating original activation addresses from the decoded activation addresses and the decoded weight addresses. Moreover, the method may include matching the original activation addresses to a mapping dictionary to generate encoded activation addresses for the current layer.

Abstract: According to one embodiment, a first set of features is received, where each of the features in the first set being associated with a predetermined category. A bloom filter is applied to the first set of features to generate a second set of features. A neural network model is trained by applying the second set of features to a first layer of nodes of the neural network model to generate an output, the neural network model including a plurality of layers of nodes coupled to each other via a connection. The output of the neural network model is compared with a target value associated with the predetermined category to determine whether the neural network model satisfies a predetermined condition.

Abstract: A database query comprising predicates may be received. Each predicate may operate on database columns. The database query may be determined to comprise strict operators. An upper bound neural network may be defined for calculating an adjacent upper bound and a lower bound neural network may be defined for calculating an adjacent lower bound. The upper bound neural network and the lower bound neural network may be trained using a selected value from data of a database table associated with the database query to be executed through the upper bound neural network and the lower bound neural network. The upper bound neural network and the lower bound neural network may be adjusted by passing in an expected value using an error found in expressions. The adjacent lower bound and the adjacent upper bound may be calculated in response to completion of initial training for the database columns.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks with an external memory using reinforcement learning. One of the methods includes providing an output derived from the system output portion of the neural network output as a system output in the sequence of system outputs; selecting a memory access process from a predetermined set of memory access processes for accessing the external memory from the reinforcement learning portion of the neural network output; writing and reading data from locations in the external memory in accordance with the selected memory access process using the differentiable portion of the neural network output; and combining the data read from the external memory with a next system input in the sequence of system inputs to generate a next neural network input in the sequence of neural network inputs.

Abstract: A recognition apparatus based on a deep neural network, a training apparatus and methods thereof. The deep neural network is obtained by inputting training samples comprising positive samples and negative samples into an input layer of the deep neural network and training. The apparatus includes: a judging unit configured to judge that a sample to be recognized is a suspected abnormal sample when confidences of positive sample classes in a classification result outputted by an output layer of the deep neural network are all less than a predefined threshold value. Hence, reliability of a confidence of a classification result outputted by the deep neural network may be efficiently improved.

Abstract: Systems and methods for neural machine translation are provided. In one example, a neural machine translation system translates text and comprises processors and a memory storing instructions that, when executed by at least one processor among the processors, cause the system to perform operations comprising, at least, obtaining a text as an input to a neural network system, supplementing the input text with meta information as an extra input to the neural network system, and delivering an output of the neural network system to a user as a translation of the input text, leveraging the meta information for translation.

Abstract: A method, system and computer-usable medium are disclosed for the use of a text simplification in a question answering (QA) system to improve ingestion quality and QA performance. A source corpus is processed to generate a parsed source corpus, which in turn is processed to generate a shadow corpus of simplified text. The parsed source corpus and the shadow corpus are then processed to generate derived data resources. A user query is processed to generate a set of merged candidate answer variants which are in turn processed to generate a corresponding converged feature vector representing each merged candidate answer variant. Feature values associated with each converged feature vector are then used to rank the merged candidate answer variants. A ranked set of merged candidate answer variants is then provided to the user.

Abstract: A method and device. The device determines an initial estimate of an amount of time for a generic user to read a document, wherein the initial estimate is determined based on a value of one or more parameters associated with the document. The device determines an estimate correction parameter for modifying the determined initial estimate to compensate for a particular user who will read the document, a particular context n which the particular user will read the document, or a combination thereof. The device uses the initial estimate and the estimate correction parameter to determine an estimate of the amount of time for the particular user to read the document.

Abstract: A method and a system are disclosed for determining a weight allocation in a group comprising a large plurality of items using an optimization oracle, the method comprising obtaining an indication of a plurality of data for each item of a large plurality of items; generating a covariance matrix for the plurality of data; generating a hierarchical tree structure having a plurality of clusters, each cluster having a corresponding item associated therewith, the generating comprising until there is one item associated per cluster of the hierarchical tree structure, recursively formulating an optimization problem to divide a given set of items into two different clusters, translating the formulated optimization problem into an unconstrained binary optimization problem, providing an indication of the unconstrained binary optimization problem to an optimization oracle, receiving an indication of at least one solution from the optimization oracle, assigning a cluster to each item of the given set of items using the a

Abstract: A method for determining occupancy of a facility is provided.

Abstract: An apparatus for determining occupancy of a facility is provided. The apparatus includes a facility model processor and a global model module. The facility model processor is configured to generate occupancy components for the facility by processing a first data set comprising energy consumption and outside temperature data for the facility, the energy consumption and outside temperature data taken at a prescribed time increment over a first plurality of days, and is configured to generate a normalized first data set by employing the occupancy components to remove effects of occupancy of the facility from the first data set. The occupancy components include: a lower bound of energy consumption as a function of outside temperature; a normalized occupancy profile component as a function of the prescribed time increment; a marginal energy consumption component as a function of outside temperature; and a daily occupancy level component for each of the first plurality of days.

Abstract: A situational awareness system providing geographic, personal characteristic and subject matter of interest awareness using information obtained from location reporting mobile devices. The system displays personal characteristic information for target member segments defined with a great deal of specificity, which may be followed by online queries to the target member segment. Member participation in online queries may require interested members to “opt-in” to limit query distribution to members that have indicated an interest in responding to online queries on the particular subject matter of interest. This increases the efficiency and effectiveness of the online queries, while reducing the cost and imposition on members who are not be interested in participating.

Abstract: An article of manufacture for predicting the popularity of media may include at least one processor readable storage medium and instructions stored on the at least one medium. The instructions may be configured to be readable from the at least one medium by at least one processor and may thereby cause the at least one processor to operate so as to determine a publication time associated with digital media, set a predetermined time threshold based on the publication time, analyze the digital media to determine at least one feature associated with the digital media, compile the determined at least one feature, predict a popularity associated with the digital media based on the compiled at least one feature, and output the predicted popularity. The at least one feature associated with the digital media may be determined before the predetermined time threshold.

Abstract: Disclosed is a system (and process) for determining the accuracy of computerized tasks in a task batch. The system calculates a number of reviews to assess the accuracy of a task based on a source accuracy and a reviewer accuracy. The source accuracy is based factors calculated by a predictive model, the factors including a historical accuracy of an authoring user. The reviewer accuracy is based on a true positive rate and a true negative rate of one or more reviewers of the task batch. The system transmits sourced tasks to a same number of reviewers. The system collects reviews and assesses if the task passes review based on the collected number of reviews.

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

Abstract: A digital marking processing apparatus includes a central control unit, and a processing quality prediction unit, a processing unit and a marking unit which are respectively connected with the central control unit electrically. The processing quality prediction unit can implement a virtual processing quality prediction method to predict the processing quality of the workpiece, output an accurate data of quality to the central control unit and generate tool path for the processing unit to process the workpiece. The central control unit is able to compile the data of quality from the processing quality in prediction unit into file information, so that the marking unit can then utilize the file information to correspondingly mark barcode or other digital pattern on the workpiece, which facilitate workpiece management and information disclosure.

Abstract: The various implementations described herein include systems and methods for determining that media items are currently being presented. In one aspect, a method performed at a client device includes: (1) identifying a media item potentially being presented on an output device coupled to the client device; (2) calculating a level of confidence that the identified media item is currently being presented to the user; (3) subsequent to the calculating, determining that at least one predetermined event associated with user interaction with the media item has occurred; and (4) based on the at least one predetermined event, increasing or decreasing the level of confidence that the identified media item is currently being presented to the user.

Abstract: Systems, methods, and non-transitory computer-readable media can acquire a set of individual time series associated with a set of users. Each of the individual time series can be associated with a respective user out of the set of the users. A plurality of variables represented via the set of individual time series can be selected. The plurality of variables can include at least a first variable and a second variable. One or more regression techniques can be applied to at least the first variable and the second variable. A set of sensitivity metrics for the set of users can be determined based on the one or more regression techniques. A respective sensitivity metric out of the set of sensitivity metrics can be determined for each of the users.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining retraining predictive models. One of the methods includes maintaining, by a computer system of an enterprise, one or more predictive models. The computer system receives operational data and uses each of the one or more predictive models to generate predictions using the received operational data. An indication of a systemic change in the computer system is received. The method includes determining that one or more retraining rules specify that at least one of the one or more predictive models should be retrained due to the systemic change, and in response, obtaining updated training data and retraining the predictive model using the updated training data.

Abstract: A plurality of binary support vector machines (SVM) are trained to vote between two of a plurality of classes. Each of the binary SVMs vote between a different pair of the plurality of classes. A set of validated samples is input to each of the binary SVMs. The set of validated samples includes samples of each of the classes. Likelihood estimates are determined based on votes of the binary SVMs. A prior probability distribution (prior) is assigned for each of the classes. A posterior probability is calculated for each of the classes based on the determined likelihood estimates and the assigned priors.

Abstract: The present invention relates to a method and an apparatus for estimating multiple-ranking using pairwise comparative data, the method according to one embodiment comprising the steps of: inputting information for pairwise comparative data as input; and generating multiple-ranking data on the basis of the input data, wherein the input data comprises a collection of a plurality of respondents, a collection of a plurality of items, and a collection of a plurality of pairwise comparative data for the plurality of items, and each item of pairwise comparative data in the plurality of same is data indicating the preference between two items determined in accordance with an evaluation criterion from among a plurality of evaluation criteria.

Abstract: Roughly described, an evolutionary data mining system includes at least two processing units, each having a pool of candidate individuals in which each candidate individual has a fitness estimate and experience level. A first processing unit tests candidate individuals against training data, updates an individual's experience level, and assigns each candidate to one of multiple layers of the candidate pool based on the individual's experience level. Individuals within the same layer of the same pool compete with each other to remain candidates. The first processing unit selects a set of candidates to retain based on the relative novelty of their responses to the training data. The first processing unit reports successful individuals to the second processing unit, and receives individuals for further testing from the second processing unit. The second processing unit selects individuals to retain based on their fitness estimate.

Abstract: Technologies are described herein for modifying the modality of a computing device based upon a user's brain activity. A machine learning classifier is trained using data that identifies a modality for operating a computing device and data identifying brain activity of a user of the computing device. Once trained, the machine learning classifier can select a mode of operation for the computing device based upon a user's current brain activity and, potentially, other biological data. The computing device can then be operated in accordance with the selected modality. An application programming interface can also expose an interface through which an operating system and application programs executing on the computing device can obtain data identifying the modality selected by the machine learning classifier. Through the use of this data, the operating system and application programs can modify their mode of operation to be most suitable for the user's current mental state.

Abstract: A computing device determines an SVDD to identify an outlier in a dataset. First and second sets of observation vectors of a predefined sample size are randomly selected from a training dataset. First and second optimal values are computed using the first and second observation vectors to define a first set of support vectors and a second set of support vectors. A third optimal value is computed using the first set of support vectors updated to include the second set of support vectors to define a third set of support vectors. Whether or not a stop condition is satisfied is determined by comparing a computed value to a stop criterion. When the stop condition is not satisfied, the first set of support vectors is defined as the third set of support vectors, and operations are repeated until the stop condition is satisfied. The third set of support vectors is output.

Abstract: An optimized method, system, and apparatus for determining optimal DQS delay for DDR memory interfaces are disclosed. The method performs data eye training in a two dimensional space with time delay value as x-axis and reference voltage (Vref) as y-axis to determine a rectangular data eye within an overall data eye with Vref margin.

Abstract: A system includes a learning object storing section that stores objects to be learned, a learning result storing section that stores learning results, and a control section connected to an input section. The control section computes a principal component coefficient vector of a first feature vector of an object to be processed that is designated by the input section, computes a principal component coefficient vector of a second feature vector using a principal component basis vector stored in the learning result storing section, and computes the second feature vector of the object to be processed using the principal component coefficient vector of the second feature vector.

Abstract: An architecture and associated techniques of an apparatus for hardware accelerated machine learning are disclosed. The architecture features multiple memory banks storing tensor data. The tensor data may be concurrently fetched by a number of execution units working in parallel. Each operational unit supports an instruction set specific to certain primitive operations for machine learning. An instruction decoder is employed to decode a machine learning instruction and reveal one or more of the primitive operations to be performed by the execution units, as well as the memory addresses of the operands of the primitive operations as stored in the memory banks. The primitive operations, upon performed or executed by the execution units, may generate some output that can be saved into the memory banks. The fetching of the operands and the saving of the output may involve permutation and duplication of the data elements involved.

Abstract: A software solution for managing, sorting and ranking lists of tasks and integrating task and time management, such that tasks can be automatically or manually assigned to specified time blocks. Users can monitor the relationship between volume of tasks and available time in which to complete them. The solution and method can be applied to individual task lists as well as to the management of time across multiple projects and can be employed either by individuals or by collaborative groups. The solution and method employ multiple filters, sorts and handling rules to embody users' personal planning preferences. It can be used to manage task lists both with and without employing a scheduling component. The solution can interoperate with existing computer- and web-based calendar software and can use third-party calendar clients to display its output and to accept input from users.

Abstract: A method for managing a network of napping pods includes a back-end system for storing a napping pod database. A plurality of napping pods is registered with the back-end system, wherein an address is stored for each of the plurality of napping pods. Customers can search for a plurality of available pods from the plurality of napping pods through a software application. The plurality of available pods can be filtered by proximity to the customer, price, rating, and other factors. The customer can then select a specific pod from the plurality of available pods and submit a pod reservation for the specific pod. The back-end system then reserves the specific pod and sends a notification to the owner of the specific pod.