Abstract: An apparatus, system and method are provided for facilitating commerce. A set of images of a first item and information associated with the first item is received and analyzed to determine at least one corresponding item category for the first item. A query for a second item is further received. The second item is determined to correspond to one of the at least one item category. In response to determining that the second item corresponds to the one of the determined at least one item category, at least one of the set of images of the first item and at least a portion of the information associated with the first item is transmitted to be presented to the second user on a display of the second device.

Abstract: Methods, systems, and computer program products are provided for processing a request regarding relationships among instances of entities. A graphical representation of instances of entities is generated and includes one or more source nodes, each representing an instance of an input entity of a request, and one or more related nodes, each representing an instance of a second entity related to one or more corresponding instances of the input entity and associated with a corresponding confidence score for the relationship. Each of the one or more related nodes associated with a confidence score satisfying a threshold is identified. One or more supplemental nodes are added to the graphical representation, each of which represents a corresponding instance of a third entity with a relationship to a corresponding instance of the second entity. The graphical representation is traversed to identify relationships between instances of entities and produce results for the request.

Abstract: During online shopping, customer buying decision varies based on conditions at the time of logging such as product availability, competitor price of the product, presence of promotion, delivery options such as number of days to deliver, availability of free delivery, and availability of pay on delivery and customer review ratings. Customer shifts from one product to other product based on the options available at real time and accordingly demand of a product is transferred to other product. The method and system disclosed provides dynamic demand transfer values that are specific to a customer for available options at the time of login and it provides the list of ideal products to be displayed at the time of customer login. The method utilizes a suitable data format to apply machine learning based approach for estimating DT, wherein training data for ML captures plurality of sales drivers affecting customer decision during online retailing.

Abstract: There is provided a method of generating a machine learning (ML) model ensemble for computing likelihood of an entity failing to meet a target parameter, comprising: training ML-sub-models that each output sub-values for an input of raw data elements, training a principal ML model that outputs a value of an entity parameter corresponding to the target parameter for an input of the sub-values, using a training dataset including for sample entities, the ML-sub-values and corresponding entity parameters, inputting raw data elements associated with the entity into the ML-sub-models to obtain respective sub-values, in iterations: computing simulated adjustments to the sub-values to generate adjusted sub-values that are inputted into the principal ML model to obtain simulated values for the entity parameter, and generating a risk classifier that generates a likelihood of the entity failing to meet the target parameter according to an analysis of the simulated values for the entity parameter.

Abstract: An exemplary computing environment having a DNN module can maintain one or more bandwidth throttling mechanisms. Illustratively, a first throttling mechanism can specify the number of cycles to wait between transactions on a cooperating fabric component (e.g., data bus). Illustratively, a second throttling mechanism can be a transaction count limiter that operatively sets a threshold of a number of transactions to be processed during a given transaction sequence and limits the number of transactions such as multiple transactions in flight to not exceed the set threshold. In an illustrative operation, in executing these two exemplary calculated throttling parameters, the average bandwidth usage and the peak bandwidth usage can be limited. Operatively, with this fabric bandwidth control, the processing units of the DNN are optimized to process data across each transaction cycle resulting in enhanced processing and lower power consumption.

Abstract: A server comprising a processor circuit and a database may receive address book data comprising information associated with at least one contact from a communication device via a network. The processor circuit may identify information associated with the at least one contact in the database and/or from public data. The processor circuit may add the identified information to the address book data. The processor circuit may store the address book data with the added information in the database and send the added information with or without the address book data to the communication device via the network.

Abstract: A vehicle ranging system includes a ranging scanner, a ranging data processor, and an object classifier. The ranging scanner may be configured to generate a series of ranging data. The ranging data processor may be configured to apply a scale invariant ranging template to the series of ranging data and determine a difference between the ranging template and the series of ranging data. The object classifier may be configured to identify an object corresponding to the series of ranging data based on the difference between the ranging template and the series of ranging data.

Abstract: Machine learning classification models which are robust against label noise are provided. Noise may be modelled explicitly by modelling “label flips”, where incorrect binary labels are “flipped” relative to their ground truth value. Distributions of label flips may be modelled as prior and posterior distributions in a flexible architecture for machine learning systems. An arbitrary classification model may be provided within the system. The classification model is made more robust to label noise by operation of the prior and posterior distributions. Particular prior and approximating posterior distributions are disclosed.

Abstract: A method of on-vehicle active learning for training a perception network of an autonomous vehicle is provided. The method includes steps of: an on-vehicle active learning device, (a) if a driving video and sensing information are acquired from a camera and sensors on an autonomous vehicle, inputting frames of the driving video and the sensing information into a scene code assigning module to generate scene codes including information on scenes in the frames and on driving events; and (b) at least one of selecting a part of the frames, whose object detection information satisfies a condition, as specific frames by using the scene codes and the object detection information and selecting a part of the frames, matching a training policy, as the specific frames by using the scene codes and the object detection information, and storing the specific frames and specific scene codes in a frame storing part.

Abstract: Embodiments of the invention provide a method comprising creating a structural description for at least one neurosynaptic core circuit. Each core circuit comprises an interconnect network including plural electronic synapses for interconnecting one or more electronic neurons with one or more electronic axons. The structural description defines a desired neuronal activity for the core circuits. The desired neuronal activity is simulated by programming the core circuits with the structural description. The structural description controls routing of neuronal firing events for the core circuits.

Abstract: Embodiments of the present invention provide an artificial neural network system for improved machine learning, feature pattern extraction and output labeling. The system comprises a first spiking neural network and a second spiking neural network. The first spiking neural network is configured to spontaneously learn complex, temporally overlapping features arising in an input pattern stream. Competitive learning is implemented as Spike Timing Dependent Plasticity with lateral inhibition in the first spiking neural network. The second spiking neural network is connected with the first spiking neural network through dynamic synapses, and is trained to interpret and label the output data of the first spiking neural network. Additionally, the output of the second spiking neural network is transmitted to a computing device, such as a CPU for post processing.

Abstract: A training task optimization system includes a processor. The processor is configured to receive training environment information of a training task. The training environment information at least carries information corresponding to training samples in the training task. The processor is configured to calculate a memory distribution for the training task based on memory factors, the training samples and a neural network, and select a mini-batch size that is fit to the memory distribution. In response to the training environment information, the processor is configured to output the mini-batch size for execution of the training task.

Abstract: Disclosed is a method and system for detecting malicious entities and malicious behavior in a time evolving network via a graph framework by modeling activity in a network graph representing associations between entities. The system utilizes classification methods to give score predictions indicative of a degree of suspected maliciousness, and presents a unified graph inference method for surfacing previously undetected malicious entities that utilizes both the structure and behavioral features to detect malicious entities.

Abstract: In one embodiment, a method, apparatus, and system may predict behavior of environmental objects using machine learning at an autonomous driving vehicle (ADV). One or more yield/overtake decisions are made with respect to one or more objects in the ADV's surrounding environment using a data processing architecture comprising at least a first, a second, and a third neural networks, the first, the second, and the third neural networks having been trained with a training data set. Driving signals are generated based at least in part on the yield/overtake decisions to control operations of the ADV.

Abstract: A method, computer program product, and computer system, for receiving a first set of ground truth instances from a first source. A second set of ground truth instances may be received from a second source. The first and second sets of ground truth instances may be weighed differently based on a level of trust associated with each of the first and second sources. The weighted first and second sets of ground truth instances may be applied in a machine learning task executed by a computer.

Abstract: A system and method for for grouping of similar image classes for image recognition is provided. The invention comprises extracting one or more features from multiple classes of images for determining a correlation value between each of the multiple classes of images based on assessment of the extracted features of each one of the classes of images with respect to other classes of images in the multiple classes of images. Further, the one class of image is grouped with the other class of image in the multiple classes of images to form one or more groups of super-classes of similar class of images based on analysis of the determined correlation values with respect to a pre-determined threshold value. An input image is recognized based on the formed groups of super-classes followed by sub-class classification of the images.

Abstract: Various embodiments are provided for managing and interacting with a conversational agent in an internet of things (IoT) environment by a processor. One or more graph patterns may be extracted from a knowledge graph generated from a sensor list containing one or more IoT sensors. One or more conversation patterns and conversation dialogs may be learned from the sensor list and graph patterns. The conversational agent may be automatically configured with the one or more graph patterns, the one or more conversational patterns and conversation dialogs, sensor data and analytics from the one or more IoT sensors according to one or more queries to engage in a conversation dialog with a user.

Abstract: Systems and methods for realigning streams of neuron outputs are provided. An example method may include generating, by a processing unit, neuron outputs including at least a first neuron output and a second neuron output, generating, by at least one further processing unit, further neuron outputs including at least a further first neuron output and a further second neuron output, receiving, by a synchronization module communicatively coupled to the processing unit and the further processing unit, the neuron outputs, wherein the neuron outputs and the further neuron outputs are received in an arbitrary order, and ordering, by the synchronization module, the first neuron output, the further first neuron output, the second neuron output and the further second neuron output according to a further order, the further order being different from the arbitrary order.

Abstract: A display apparatus includes: a display; a memory storing one or more instructions; and a processor configured to execute the one or more instructions stored in the memory to: select an image to be displayed on the display, classify a category of the image, obtain semantic information of the image according to the category of the image, by using at least one neural network, obtain emotion information corresponding to the image by analyzing the image; determine an audio to be output together with the image, based on the semantic information of the image and the emotion information corresponding to the image, and output the audio while displaying the image.

Abstract: A method, computer readable medium, and system are disclosed for performing spatiotemporal filtering. The method includes identifying image data to be rendered, reconstructing the image data to create reconstructed image data, utilizing a filter including a neural network having one or more skip connections and one or more recurrent layers, and returning the reconstructed image data.

Abstract: A system for online advertising. The system may include at least one memory unit for storing instructions and at least one processor configured to execute the instructions to perform operations. The operations may include receiving a plurality of creatives for a webpage published to a viewer and resulting in a viewer experience; measuring, based on the viewer experience, a result including conversion rates for a plurality of variants of the webpage; determining confidence intervals in association with the conversion rates; dynamically comparing the received creatives, the conversion rates, and the determined confidence intervals of the variants; automatically analyzing, based on the comparison, the variants to select a winning webpage, the winning webpage exceeding a computed threshold; and automatically adjusting online traffic such that the selected winning webpage is displayed more frequently than other webpages.

Abstract: A system and method for associating an address for an object is disclosed. A smart address server is provided with a processor and memory. The smart address server is configured to receive a geo location for the object, the geo location indicative of the latitude and longitude of the location of the object. The geo location of the object is converted into a zone, sector, and target portion to generate a ZST location indicator for the object. An elevation portion is appended to the ZST location, to generate a ZSTE location indicator for the object. An object ID is assigned to the object and the object ID is appended to the ZSTE location indicator to generate a smart address for the object. The generated smart address is stored in a data store.

Abstract: The present disclosure is directed to optimizing transfer learning for neural networks by creating a dynamic transfer network configuration through gated architecture. In some embodiments, transfer learning implements multiple parameter sharing schemes across a source task and a target task. The gating architecture can learn the optimal parameter sharing schemes as the neural network is trained. In some embodiments, the system can be used in named entity recognition applications where the training data is limited.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for summarizing a call. One of the methods includes generating text corresponding to processing audio produced during an interaction between two participants by executing natural language processing logic. The method includes identifying one or more topics by providing the generated text to a machine learning system, the machine learning system trained to identify topics based on text. The method also includes generating a summary of the interaction based on the one or more topics and the text.

Abstract: According to the embodiments of the present disclosure, there are provided a method and device for identifying an object and a computer readable storage medium. The method includes: determining candidate categories of the object using a first neural network; and determining, in response to the candidate categories of the object having been determined, a category of the object using second neural networks respectively corresponding to the candidate categories. The device includes: a processor; a memory having stored thereon instructions which, when executed by the processor, cause the processor to: determine candidate categories of the object using a first neural network; and determine, in response to the candidate categories of the object having been determined, a category of the object using second neural networks respectively corresponding to the candidate categories.

Abstract: Embodiments of the innovation relate to a method and apparatus for utilizing graph theory and clustering analysis to build relationships among applications and corresponding and/or dependent virtualization, cloud and datacenter infrastructure components and to derive the definition of application boundaries and corresponding infrastructure components. This allows a system administrator to manage and address complex issues within a computer infrastructure, such as availability, service level deliverable/guarantees, performance, and resource optimization.

Abstract: Techniques for employing user interfaces to output information indicative of events occurring in an inventory facility, and receive feedback from a human regarding the events are described herein. In one implementation, an event may take place in an inventory facility, such as a customer taking an item from an inventory location, returning an item to an inventory location, and so forth. An automated system of an inventory management system may process sensor data collected by sensors in the inventory facility to determine details of the event. In some examples, the inventory management system is unable to determine with a high level of confidence what occurred during the event. The inventory management system may provide the sensor data to a human associate through an associate interface, and receive input regarding details of the event from the human associate through the associate interface.

Abstract: A method for detecting and localizing a fault in a system under test (SUT) includes generating an initial set of test cases using combinatorics test design. The method further includes determining, based at least in part on a first set of execution results, a set of failing test cases. The method further includes determining, based on a machine learning model, a subset of false positives from the failing test cases. The method further includes generating a set of new test cases from a selected failing test case that is not in the subset of false positives. The method further includes executing the set of new test cases to obtain a second set of execution results, and localizing the fault based at least in part on the second set of execution results.

Abstract: Techniques are disclosed for generating tutorial recommendations to users of image editing applications, based on image content. A methodology implementing the techniques according to an embodiment includes using neural networks configured to determine subject matter of a user provided image and to identify objects in the image. The method also includes selecting one or more proposed tutorials from a database of tutorials. The database is indexed by tutorial subject matter and tutorial object content, and the selection is based on a matching of the determined subject matter to the tutorial subject matter and a matching of the identified objects to the tutorial object content. The method further includes calculating an effectiveness score associated with each of the proposed tutorials, the effectiveness score based on application of the proposed tutorial to the image. The method further includes sorting the proposed tutorials for recommendation to the user based on the effectiveness scores.

Abstract: A computer system uses a pool of predefined functions and pre-trained networks to accelerate the process of building a large neural network or building a combination of (i) an ensemble of other machine learning systems with (ii) a deep neural network. Copies of a predefined function node or network may be placed in multiple locations in a network being built. In building a neural network using a pool of predefined networks, the computer system only needs to decide the relative location of each copy of a predefined network or function. The location may be determined by (i) the connections to a predefined network from source nodes and (ii) the connections from a predefined network to nodes in an upper network. The computer system may perform an iterative process of selecting trial locations for connecting arcs and evaluating the connections to choose the best ones.

Abstract: Methods, systems, and devices are described for identifying images which may be of interest to a user based on their current geographic location. In some embodiments, a check is first performed to determine if the current geographic location is a location-of-interest. Images are searched that are in geographical proximity to the current geographic location of the user to identify images-of-interest. The images-of-interest may be designated in part based on actions taken by subjects having had interactions with the images. The user is notified based on the discovery of one or more images-of-interest. The one or more images-of-interest may be presented to the user through the use of map overlays and/or augmented reality techniques.

Abstract: A collection of techniques is disclosed to allow for the detection of malware that leverages pattern recognition and machine learning to effectively provide “content-less” malware detection, i.e., detecting a process as being an ‘anomaly’ not based on its particular content, but instead based on comparisons of its behavior to known (and characterized) ‘trusted’ application behaviors, i.e., the trusted applications' “phenotypes” and/or the phenotypes of known malware applications. By analyzing the patterns of normal behavior performed by trusted applications as well as malware applications, one can build a set of sophisticated, content-agnostic behavioral models (i.e.

Abstract: The disclosed technology teaches a deep end-to-end speech recognition model, including using multi-objective learning criteria to train a deep end-to-end speech recognition model on training data comprising speech samples temporally labeled with ground truth transcriptions.

Abstract: A detector includes a housing having a window. A detection engine is configured to detect an event external to the housing. A battery is connected to the detection engine and configured to supply power thereto. A solar cell is configured to generate electricity when exposed to a light received through the window. The solar cell is arranged in the housing and electrically connected to the battery.

Abstract: A data processing apparatus may utilize an artificial neuron network configured to reduce dimensionality of input data using a sparse transformation configured using receptive field structure of network units. Output of the network may be analyzed for temporally persistency that is characterized by similarity matrix. Elements of the matrix may be incremented when present activity unit activity at a preceding frame. The similarity matrix may be partitioned based on a distance measure for a given element of the matrix and its closest neighbors. Stability of learning of temporally proximal patterns may be greatly improved as the similarity matrix is learned independently of the partitioning operation. Partitioning of the similarity matrix using the methodology of the disclosure may be performed online, e.g., contemporaneously with the encoding and/or similarity matrix construction, thereby enabling learning of new features in the input data.

Abstract: In one aspect, a computerized method useful for handwriting recognition (HWR) on physical mail envelopes addressed to a user includes the step of scanning a physical mail item to obtain a digital image of the address-side of the physical mail item. The method includes the step of identifying that at least one of a return address region or a recipient address region of the address-side of the physical mail item. The method includes the step of determining that the at least one of the return address region or a recipient address region comprises a handwritten text. The method includes the step of providing a data store of known senders to the recipient address. The data store of known senders comprises a data store of return address information in a known sender handwriting samples and a data store of receiver address information in the known sender handwriting samples. The method includes the step of providing a data store of a receiver's identity and address.

Abstract: Embodiments of the disclosure provide systems and methods for correcting a high-definition map. The system may include a communication interface configured to receive point cloud data of a scene captured by a LiDAR. The system may further include a storage configured to store the point cloud data, and at least one processor. The at least one processor may be configured to detect at least one obstructing object from the point cloud data, and position at least one hole in the point cloud data caused by the at least one obstructing object. The at least one processor is further configured to estimate non-obstructed point cloud data for the at least one hole as if the scene was captured without the at least one obstructing object, and correct the high-definition map by repairing the received point cloud data with the non-obstructed point cloud data.

Abstract: A device may receive information associated with an entity. The information may include a first resource and a second resource. The first resource may be associated with a first file type, and the second resource may be associated with a second file type that is different than the first file type. The first resource may be associated with a first source, and the second resource may be associated with a second source that is different than the first source. The device may extract a plurality of attributes associated with the entity based on the information. The device may implement a natural language processing technique to extract the plurality of attributes. The device may associate the plurality of attributes with a plurality of elements based on extracting the plurality of attributes. The device may provide information that identifies the plurality of elements and the plurality of attributes to permit and/or cause an action to be performed.

Abstract: An application may be integrated into an operating system component of a mobile computing platform. The application communicates registration information to an extension mechanism within the mobile computing platform. The extension mechanism registers the application and modifies the operating system component to provide interoperability with the application based on the registration information. The extension mechanism couples the operating system component with content from a counterpart application to the application, which corresponds to information presented on the operating system component.

Abstract: Implementations of the present specification disclose a method for generating word vectors, apparatus, and device. The method includes: obtaining words by segmenting a corpus; establishing a feature vector of each obtained word based on n-ary characters; training a convolutional neural network based on the feature vectors of the obtained words and the feature vectors of context words associated with each obtained word in the corpus; and generating a word vector for each obtained word based on the feature vector of the obtained word and the trained convolutional neural network.

Abstract: Provided are systems and method for identifying, generating and displaying a trend of one or more medical waveform related features.

Abstract: A method for utilizing a millimeter wave drilling apparatus is provided. The method can include monitoring a permeability of a first portion of a borehole of a well while forming the borehole using a drilling apparatus including a drill bit for mechanical removal of material from within the borehole. The method can also include determining to utilize a millimeter wave drilling apparatus including a wave guide configured for insertion into the borehole based at least on the permeability of the borehole falling below a permeability threshold value. The method can further include forming a second portion of the borehole utilizing the millimeter wave drilling apparatus in response to the determining. A method for controlling the downhole pressure of a well is also provided. Related systems performing the methods are also provided.

Abstract: The disclosed computer-implemented method for cross-product malware categorization may include accessing computer readable media storing an incomplete feature dataset and an incomplete label dataset, determining a correlation between the plurality of features and the plurality of malware labels, and constructing at least one of a complete feature dataset based on the incomplete feature dataset and the correlation and a complete label dataset based on the incomplete label dataset and the correlation. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Mechanisms are provided for obfuscating training of trained cognitive model logic. The mechanisms receive input data for classification into one or more classes in a plurality of predefined classes as part of a cognitive operation of the cognitive system. The input data is processed by applying a trained cognitive model to the input data to generate an output vector having values for each of the plurality of predefined classes. A perturbation insertion engine modifies the output vector by inserting a perturbation in a function associated with generating the output vector, to thereby generate a modified output vector. The modified output vector is then output. The perturbation modifies the one or more values to obfuscate the trained configuration of the trained cognitive model logic while maintaining accuracy of classification of the input data.

Abstract: Systems and methods generate a segmentation network for image segmentation using global optimization. A method for automatic generation of at least one segmentation network includes providing an initial set of hyperparameters to construct a segmentation network. The hyperparameters define operations for a set of block structures and connections between the block structures. The segmentation network is trained using a first set of images with ground truth. An objective function value for the trained segmentation network is generated using a second set of images having ground truth. The set of hyperparameters is updated by performing a derivative-free optimization algorithm on the objective function value to construct an updated segmentation network. The training of the segmentation network, the generating of the objective function, and the updating of the set of hyperparameters for the updated segmentation network are iterated to generate a network architecture for the segmentation network.

Abstract: An online system maintains machine learning models that determine risk scores for content items indicating likelihoods of content items violating content policies associated with the machine learning models. When the online system obtains an additional content policy, the online system applies a maintained machine learning model to a set including content items previously identified as violating or not violating the additional content policy. The online system maps the risk scores determined for content items of the set to likelihoods of violating the additional content policy based on the identifications of content times in the set violating or not violating the additional content policy. Subsequently, the online system applies the maintained machine learning model to content items and determines likelihoods of the content items violating the additional content policy based on the mapping of risk scores to likelihood of violating the additional content policy.

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

Abstract: A system to detect a feature in an input image comprising a processor to evaluate a model including: four layers including: a supragranular layer, a granular layer, a first infragranular layer, and a second infragranular layer, each of the layers including a base connection structure including: an excitatory layer including a excitatory neurons arranged in a two dimensional grid; and an inhibitory layer including a inhibitory neurons arranged in a two dimensional grid; within-layer connections between the neurons of each layer in accordance with a Gaussian distribution; between-layer connections between the neurons of different layers, the probability of a neuron of a first layer of the different layers to a neuron of a second layer of the different layers in accordance with a uniform distribution; and input connections from lateral geniculate nucleus (LGN) neurons of an input LGN layer to the granular layer in accordance with a uniform distribution.

Abstract: An object identification method is disclosed. The method includes training a first neural network for a first set of conditions regarding a first plurality of objects, training a second neural network for a second set of conditions regarding a second plurality of objects, receiving a plurality of target images associated with a third set of conditions in which to identify objects, analyzing the plurality of target images using the first and second neural networks to identify objects in the plurality of target images resulting in object identification information, and selecting the first neural network or the second neural network as a preferred neural network for the third set of conditions based on an analysis of the object identification information.

Abstract: Neural network systems are provided that comprise one or more neural networks. The first neural network can comprise a convolutional neural network (CNN) long short-term memory (LSTM) architecture for receiving a primary data set comprising text messages and output a primary data structure comprising a text pattern-based feature. The second neural network can comprise a CNN architecture for receiving a secondary data sets derived from the primary data set and output a plurality of secondary data structures. The third neural network can combine the data structures to produce a combined data structure, and then process it to produce a categorized data structure comprising the text messages assigned to targets. The primary data set can comprise hate speech and the categorized data structure can comprise target categories, for example, hate targets. Methods of operating neural network systems and computer program products for performing such methods are also provided.