Abstract: Methods and systems are disclosed for receiving and processing data analysis expressions. A particular method includes receiving a data analysis expression at a pivot table of a spreadsheet. The data analysis expression is executed for a particular cell of the pivot table by determining a context associated with the particular cell, calculating a value of the data analysis expression based on the context, and outputting the calculated value at the particular cell.

Abstract: Methods and systems for publishing text files containing mathematical expressions to a web-ready file are disclosed. The web-ready file may be used for generating a web application, creating an offline file, etc. Template documents are received as text files having a template structure comprising a plurality of identifier fields and content fields, each content field associated with a respective identifier field, and data in at least one content field comprises a mathematical expression. The mathematical expression is converted to a first web language format within the template document. The template document is exported to a second web language format. The web-ready file is generated from the exported file by assembling contents of the exported file in accordance with the plurality of content fields and associated respective identifier fields defined in the template document.

Abstract: Various methods, apparatuses/systems, and media for real-time template authoring and execution are provided. A receiver receives a request to create a real-time document template. A processor implements a template authoring work flow process by leveraging native tools to create the real-time document template based on the received request; creates a real-time document based on the real-time document template; triggers an automatic verification process or a manual verification process in response to creating the real-time document; and validates the real-time document for execution after completing the automatic verification process or the manual verification process.

Abstract: Methods and systems for using machine learning models to organize and select modular components for user interface templates. For example, the methods and systems may use machine learning and/or artificial intelligence to generate user interfaces with a preferred number of user input fields and/or pages, in which the preferred number both satisfies any requirements for the user interface (e.g., information, categories of information, etc.), and ensures that all privacy concerns are met.

Abstract: A computer-implemented method, non-transitory computer readable storage medium, and computing device for visualizing entities and relations within a file. The content of at least one portion of a file is parsed by natural language processing (NLP). A frequency is determined of one or more entities within the at least one portion of the file parsed by NLP. Visualization data is generated of at least one entity of the one or more entities in the file, and is output for display.

Abstract: A method for performing sentiment analysis on Arabic text may be described. Training text data may be preprocessed by removing non-Arabic characters, numbers, control characters or graphics. Since Arabic words may include the same letters but written in a different format, an embodiment may identify common letters and unify them in order to remove or avoid duplicates. An annotator may label portions of the data, such as words, terms, or phrases, as positive, negative. A lexicon may be formed based on the labeled training data. The bag-of-phrases may be formed from the training text data, which may be used to analyze the targeted data for sentiment. Based on the distribution of words or phrases, a sentiment may be formed indicating a sentiment of each portion of the target data.

Abstract: Methods, and systems for named entity recognition and disambiguation. One of the methods includes: receiving a document; extracting a plurality of mentions within the document; clustering the plurality of mentions to produce at least one cluster; identifying a plurality of candidate entities for the cluster; pairing each candidate entity and the cluster to create a plurality of candidate entity-cluster pairings; generating features for each candidate entity-cluster pairing; and selecting an entity, from the plurality of candidate entities, for the cluster based at least in part on features of a candidate entity pairing.

Abstract: Systems and methods are described for generating learning roadmaps from unstructured information. The systems and methods may provide for extracting a plurality of named entities from one or more corpora of information, constructing a graph based on the named entities, inducing a subgraph from the graph based on a target named entity, wherein the subgraph includes a subset of the named entities, ordering the subset of the named entities based on the subgraph, and generating a learning roadmap for the target named entity based on the ordering.

Abstract: A processor may generate a plurality of intermediate feature layers of an image using convolutional neural network (CNN) processing. For each intermediate feature layer, the processor may generate a plurality of text proposals using a region proposal network (RPN). Each text proposal may comprise a portion of the intermediate feature layer that is predicted to contain text. The processor may perform OCR processing on image data within a plurality of regions of the image to generate a text result for each region. Each region may comprise at least one of the text proposals. The processor may assemble the text results into a text string comprising the text results ordered according to a spatial order in which the plurality of regions appear within the image.

Abstract: According to some embodiments, systems and methods are provided to develop or provide common sense auto-generated explanations (CAGE) for the reasoning used by an artificial intelligence, neural network, or deep learning model to make a prediction. In some embodiments, the systems and methods use supervised fine-tuning on a language model (LM) to generate such explanations. These explanations may then be used for downstream classification.

Abstract: A semantic analysis method includes: inputting a word vector of each word in each sample sentence in a dialog flow into an encoder model, to obtain a sentence vector representing semantics of the sample sentence; inputting the sentence vector into a first decoder model corresponding to each context sentence of the sample sentence and a second decoder model corresponding to each word of the sample sentence, to obtain a first identifier corresponding to the context sentence and a second identifier corresponding to the word; obtaining a probability of the first identifier and a probability of the second identifier, and determining a value of a target function; performing parameter training on the encoder model according to the value of the target function; and inputting a word vector of each word in a test sentence into the trained encoder model, to obtain a sentence vector representing semantics of the test sentence.

Abstract: In one embodiment, a method includes receiving, from a client system associated with a first user, a first audio input. The method includes generating multiple transcriptions corresponding to the first audio input based on multiple automatic speech recognition (ASR) engines. Each ASR engine is associated with a respective domain out of multiple domains. The method includes determining, for each transcription, a combination of one or more intents and one or more slots to be associated with the transcription. The method includes selecting, by a meta-speech engine, one or more combinations of intents and slots from the multiple combinations to be associated with the first user input. The method includes generating a response to the first audio input based on the selected combinations and sending, to the client system, instructions for presenting the response to the first audio input.

Abstract: Provided are processes of balancing between exploration and optimization with knowledge discovery processes applied to unstructured data with tight interrogation budgets. In an evaluation for which features are structured (e.g., either in a structured evaluation or determined from unstructured data and provided for evaluation) a probabilistic graphical network may graph inputs of machine learning model(s) and outputs of the machine learning model(s) as graphical elements, where one or more edges or nodes, or values associated therewith, may be based on the outputs. For example, as a set of ranking entities engage an expert system during an evaluation, the expert system may determine and update a probabilistic graphical network that represents a state of the evaluation (e.g., at a point in time after one or more ranking events), or (e.g., after completion) a final state and determined scores based the inputs provided by the ranking entities.

Abstract: Embodiments of the present disclosure disclose a method and apparatus for determining a topic. A specific embodiment of the method comprises: determining a to-be-recognized sentence sequence; calculating similarities between the to-be-recognized sentence sequence and each of topic templates in a topic template set in a target area, the each of the topic templates in the topic template set corresponding to a topic in at least one topic in the target area, the topic template including a topic section sequence, and a topic section including a topic sentence sequence; and determining a topic of the to-be-recognized sentence sequence according to an associated parameter, the associated parameter including the similarities between the to-be-recognized sentence sequence and the each of the topic templates in the topic template set. This embodiment reduces labor costs during a topic segmentation.

Abstract: A gaming table includes a network analyzer. The network analyzer detects antenna reflection coefficients faster than RFID tags can be read, enabling game state information to be monitored quickly.

Abstract: An apparatus (1) for inspecting an identity document (D) is disclosed, wherein the identity document, which includes an RFID tag and a printed code, is positioned on the apparatus in a reading area (3) where a linear image sensor, which is movable by a motor-driven command, performs a scan to read the code of the document, and in which an RFID device communicates with the RFID tag, authentication of the identity document being performed by comparing the information supplied by the linear image sensor and the information supplied by the RFID device.

Abstract: A method includes: storing, in a first database, first data regarding a life cycle of a product made using a manufacturing process; scanning an encoded image on a physical object to obtain at least one parameter associated with the product; querying, using the obtained at least one parameter, the first database to obtain the first data; generating a communication that includes the first data; and sending, to a server over a network, the communication, wherein the server is configured to update a second database based on the first data.

Abstract: A method for determining a threshold for spike detection in a noisy signal in real-time is provided. The method includes estimating a current variability of noise in the noisy signal according to a measured instantaneous value and a window of previous instantaneous values using a sigma-delta control loop, determining the threshold based on the estimated variability of the noise; and repeating the steps to update the estimate of the variability of the noise and adjust the threshold in real-time as the noisy signal changes. A non-transitory computer-readable storage medium for executing the method on a processing unit, and a low-power digital system implementing the method are also provided.

Abstract: A data recognition method includes: extracting a feature map from input data based on a feature extraction layer of a data recognition model; pooling component vectors from the feature map based on a pooling layer of the data recognition model; and generating an embedding vector by recombining the component vectors based on a combination layer of the data recognition model.

Abstract: Image data is accessed. The image data includes frequency domain components. A subset of the frequency domain components is selected based on the relative importance of the frequency domain components. Only the subset of the frequency domain components is provided to an accelerator that executes a neural network to perform an artificial intelligence task using the subset of frequency domain components.

Abstract: A method of selecting data for privacy preserving machine learning comprises: storing training data from a first party, storing a machine learning model, and storing criteria from the first party or from another party. The method comprises filtering the training data to select a first part of the training data to be used to train the machine learning model and select a second part of the training data. The selecting is done by computing a measure, using the criteria, of the contribution of the data to the performance of the machine learning model.

Abstract: Providing localization data includes obtaining a first image of a scene associated with a first condition, determining one or more target conditions, and applying an appearance transfer network to the first image to obtain one or more synthesized images comprising the scene, wherein the scene is associated with the one or more target conditions in the synthesized image. A first patch is selected from the first image, wherein the first patch comprises a keypoint, and an image location is determined for the first patch. Then one or more additional patches can be obtained using the synthesized images and the image location. A descriptor network may be trained to provide localization data based on the first patch and the one or more additional patches.

Abstract: Various examples are directed to systems and methods for detecting training data for a generative model. A computer system may access generative model sample data and a first test sample. The computer system may determine whether a first generative model sample of the plurality of generative model samples is within a threshold distance of the first test sample and whether a second generative model sample of the plurality of generative model samples is within the threshold distance of the first test sample. The computer system may determine that a probability that the generative model was trained with the first test sample is greater than or equal to a threshold probability based at least in part on whether the first generative model sample is within the threshold distance of the first test sample, the determining also based at least in part on whether the second generative model sample is within the threshold distance of the first test sample.

Abstract: One or more processors may identify a missing image in the set of multi-view images. Each of the images is associated with a particular view type. The one or more processors may generate, utilizing a replacement AI model, a replacement image for the missing image in the set of multi-view images. The replacement image is generated utilizing an AI model trained to generate a replacement image using training images from two or more time-adjacent sets of images. The one or more processors may identify a duplicate image in the set of multi-view images. The one or more processors may generate, utilizing a quality generative AI model, a characteristic improved image based on the duplicate image for the set of multi-view images. The one or more processors may output the replacement image and the characteristic improved image.

Abstract: Embodiments include a method, an electronic device, and a computer program product for information processing. In an example embodiment, a method for information processing includes: acquiring, at a first device, a first feature associated with a target object; applying the first feature to a trained first model deployed at the first device to determine a first confidence coefficient, the first confidence coefficient being associated with probabilities that the first model determines the target object as a real object and as a false object; if the first confidence coefficient is lower than a first threshold confidence coefficient, sending a request for verifying the target object to a second device, the second device being deployed with a trained second model for verifying the target object, and the second model being more complex than the first model; and updating the first model based on a response to the request.

Abstract: In medicine and dentistry, image quality affects computer vision accuracy. However, some problems are more tolerant of noise depending on disease severity and radiographic obviousness. There is a need to have a noise estimation model that adapts to each specific domain. A noise estimation model is trained to output a set of domain noise estimates for an input image, each estimate indicating an impact of noise present in the input image on a particular domain, e.g. labeling of a dental feature such as a dental anatomy, pathology, or treatment. The noise estimation model is trained by processing image pairs with a set of machine learning models for a plurality of domains, the image pairs including a raw image and a modified image obtained by adding noise to the raw image. Outputs of the set of machine learning models for the raw and modified images are compared to obtain measured noise metrics. The noise estimation model processes the modified image and is trained to estimate noise metrics.

Abstract: A vehicle includes a vehicle body having a camera and at least one ego part connection. An ego part is connected to the vehicle body via the ego part connection. A collision detection system is communicatively coupled to the camera and is configured to receive a video feed from the camera. The collision detection system being configured to identify an exclusion region of each frame corresponding to the ego part, perform an object detection on a remainder of each frame, and generate a collision detection warning in response to an object being detected by the object detection system.

Abstract: A computer-implemented method of determining an explainability mask for classification of an input image by a trained neural network. The trained neural network is configured to determine the classification and classification score of the input image by determining a latent representation of the input image at an internal layer of the trained neural network. The method includes accessing the trained neural network, obtaining the input image and the latent representation thereof and initializing a mask for indicating modifications to the latent representation. The mask is updated by iteratively adjusting values of the mask to optimize an objective function, comprising i) a modification component indicating a degree of modifications indicated by the mask, and ii) a classification score component, determined by applying the indicated modifications to the latent representation and determining the classification score thereof. The mask is scaled to a spatial resolution of the input image and output.

Abstract: This disclosure relates to method and system for of dynamically annotating data or validating annotated data. The method may include receiving input data comprising a plurality of input data points. The method may further include one of: a) generating a plurality of annotations for each of the plurality of input data points using at least one of a state-label mapping model and a comparative ANN model, or b) receiving the plurality of annotations for each of the plurality of input data points from an external device or from a user, and validating the plurality of annotations using at least one of the state-label mapping model and the comparative artificial neural network (ANN) model.

Abstract: The technology described herein is directed to systems, methods, and software for indexing video. In an implementation, a method comprises identifying one or more regions of interest around target content in a frame of the video. Further, the method includes identifying, in a portion of the frame outside a region of interest, potentially empty regions adjacent to the region of interest. The method continues with identifying at least one empty region of the potentially empty regions that satisfies one or more criteria and classifying at least the one empty region as a negative sample of the target content. In some implementations, the negative sample of the target content in a set of negative samples of the target content, with which to train a machine learning model employed to identify instances of the target content.

Abstract: Embodiments of the present invention provide a computer-implemented method for performing unsupervised time-series feature learning. The method generates a set of reference time-series of random lengths, in which each length is uniformly sampled from a predetermined minimum length to a predetermined maximum length, and in which values of each reference time-series in the set are drawn from a distribution. The method generates a feature matrix for raw time-series data based on a set of computed distances between the generated set of reference time-series and the raw time-series data. The method provides the feature matrix as an input to one or more machine learning models.

Abstract: A system includes signal acquisition circuitry and a processor. The signal acquisition circuitry is configured to receive multiple intra-cardiac signals acquired by multiple electrodes of an intra-cardiac probe in a heart of a patient. The processor is configured to perform a sequence of annotation-visualization operations at subsequent times, by performing, in each operation: extracting multiple annotation values from the intra-cardiac signals, selecting a group of the intra-cardiac signals, identifying in the group one or more annotation values that are statistically deviant by more than a predefined measure of deviation, and visualizing the annotation values to a user, while omitting and refraining from visualizing the statistically deviant annotation values.

Abstract: Methods, apparatus and systems for printing an image using an array of nozzles are described. In one example aspect, a printer system includes an array of nozzles and a control device coupled to the array of nozzles. The control device is configured to determine a step size for printing a current section of an image based on a set of masks. The set of masks includes one or more masks used for printing previous sections of the image. The control device is also configured to adjust the set of masks based on a printing mode to be used for the current section of the image. The array of nozzles is configured to print the current section of the image using a combination of the adjusted set of masks.

Abstract: Provided is a management apparatus includes: an acquisition unit acquiring remaining-quantity information indicating a remaining quantity of a consumable included in the image forming apparatus; a database storing the remaining-quantity information and image forming apparatus identification information in association with each other, and storing user identification information and a usage history of the image forming apparatus by the user in association with each other; a calculation unit reading the usage history from the database based on externally-designated user identification information, calculating a predicted usage quantity of the consumable in an image forming apparatus having externally-designated apparatus identification information based on the read usage history, reading the remaining-quantity information from the database based on the externally-designated apparatus identification information, generating required information on the consumable based on the read remaining-quantity information and

Abstract: A method of manufacturing a metal product comprise forming a base region on a surface of a metal member by repeatedly scanning along a predetermined first direction while irradiating the surface of the metal member with a base laser beam over a first set of rows. The method further comprises forming a marking by repeatedly scanning along a predetermined second direction while irradiating the surface of the metal member with a marking laser beam over a second set of rows. The second direction is different from the first direction. An identification code having a predetermined pattern comprises a combination of the base region and the marking.

Abstract: Methods and systems disclosed herein may communicate information, such as alerts and notifications, to a cardholder via a color-changing transaction card. In particular, the cardholder may configure one or more thresholds that establish when a transaction card may change colors. For instance, a transaction card may change colors at, or while approaching, a first threshold. The transaction card may change colors at, or while approaching, a second threshold. When the cardholder's balance goes below the thresholds, the transaction card may revert back to the lower threshold color or the original color of the transaction card. Using the techniques described herein a financial institution may convey information to a cardholder in a way that does not require the cardholder to digitally engage with the financial institution.

Abstract: A liquid lens includes a substrate, an anti-reflection (AR) coating, and a chipless radio frequency identification (RFID) tag. The substrate includes central and peripheral portions. The AR coating is disposed on the substrate. The chipless RFID tag is disposed in the peripheral portion to uniquely identify the liquid lens.

Abstract: Systems and methods are described to enhance interactive engagement during simultaneous delivery of serial or digital content (e.g., audio, video) to a plurality of users. A machine-based awareness of the context of the content and/or one or more user reactions to the presentation of the content may be used as a basis to interrupt content delivery in order to intersperse a snippet that includes a virtual agent with an awareness of the context(s) of the content and/or the one or more user reactions. This “contextual virtual agent” (CVA) enacts actions and/or dialog based on the one or more machine-classified contexts coupled with identified interests and/or aspirations of individuals within the group of users. The CVA may also base its activities on a machine-based awareness of “future” content that has not yet been delivered to the group, but classified by natural language and/or computer vision processing.

Abstract: Systems and techniques for neuromorphic accelerator multitasking are described herein. A neuron address translation unit (NATU) may receive a spike message. Here, the spike message includes a physical neuron identifier (PNID) of a neuron causing the spike. The NATU may then translate the PNID into a network identifier (NID) and a local neuron identifier (LNID). The NATU locates synapse data based on the NID and communicates the synapse data and the LNID to an axon processor.

Abstract: It is possible to improve estimation accuracy with regard to data in which significance is attached to a relative phase. Provided is an information processing device including an estimation unit configured to estimate a status by using a neural network. The neural network includes a first complex-valued neural network to which complex data is input, a phase difference computation layer from which phase difference for each element between a plurality of sets with regard to the complex data is output, and a second complex-valued neural network from which complex data is output on the basis of the phase difference.

Abstract: According to one embodiment, a data compression apparatus includes processing circuitry. The processing circuitry generates reconstructed data by performing reconstruction processing on data. The processing circuitry generates decompressed reconstructed data by performing the reconstruction processing on decompressed data obtained by decompressing compressed data that is generated by performing compression processing on the data. The processing circuitry determines a parameter relating to a compression ratio of the data based on comparison between the reconstructed data and the decompressed reconstructed data.

Abstract: A computer-implemented method of facilitating neural network image analysis involves receiving signals representing a set of images, causing at least one neural network function to be applied to the set of images to determine at least one property confidence distribution parameter, and causing a cumulative distribution function defined at least in part by the at least one property confidence distribution parameter to be applied to a plurality of ranges, each range associated with a respective property that may be associated with the set of images, to determine a plurality of property confidences, each of the property confidences representing a confidence that the set of images should be associated with a respective one of the properties. Other methods, systems, and computer-readable media are disclosed.

Abstract: A method for constructing and training a Decentralized Migration Diagram Neutral Network (DMDNN) model for a production process, including: determining a production task input management node, distributed management nodes and a granularity of each of the distributed management nodes, and constructing a production system network; constructing network calculation nodes on each of the distributed management nodes; constructing and training a DMDNN model; and applying the trained DMDNN model in management and control of the production process.

Abstract: A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the process including obtaining a feature quantity of input data by using a feature generator, generating a first output based on the feature quantity by using a supervised learner for labeled data, generating a second output based on the feature quantity by using an unsupervised learning processing for unlabeled data, and changing a contribution ratio between a first error and a second error in a learning by the feature generator, the first error being generated from the labeled data and the first output, the second error being generated from the unlabeled data and the second output.

Abstract: Techniques are provided for replacing image assays using real world data and real word evidence RNA-seq analysis for assessing biologic pathways for identifying molecular subtypes. Systems of a methods diagnose HER2 status for a patient, by identifying discordant HER2 status result between the HER2 status from immunohistochemistry (IHC) and the HER2 status from fluorescence in-situ hybridization (FISH) and diagnosing HER2 status based gene expression data.

Abstract: An optimization calculation method includes, generating individuals of a present generation with an individual selected in a previous generation as a parent individual using an algorithm for obtaining optimum solutions while evolving a plurality of individuals for each generation, evaluating each individual of the present generation, setting a constraint condition value of the present generation based on a constraint condition value of one generation before with respect to the present generation, and a constraint condition provisional value achieved by a half or more of individuals used for generating child individuals of a next generation, determining whether an evaluation result of each individual satisfies a constraint condition value of the present generation, changing for lowering an evaluation result of individuals which do not satisfy the constraint condition value of the present generation, and selecting individuals to be solution candidates from among individuals on which the determination or change

Abstract: A system and method that takes in a data set comprising molecular structure data and properties of interest, e.g., ADMET, EC50, IC50, etc., and determines the substructures that cause or do not cause the property of interest. The substructures may then be used to filter out potentially harmful new proposed/generated molecules or create a new data set of known active/inactive substructures of a property of interest that may fulfill other obligations. The system comprises a substructure extraction module which further comprises a scaffold extraction module and a comparison module. A scaffold extraction module clusters, searches, and extracts substructures in question while a comparison module compares the bioactivity of each molecule with and without each substructure in question to determine the substructures effect on the property of interest.

Abstract: This patent specification relates to systems and methods that use networks of standardized time series and models in a generic and extensible platform. More particularly, this patent specification relates to standardizing time series and models, using standardized time series and models in a network of time series (NOTS), and creating networks of time series in a NOTS platform. In addition, this patent specification relates to use of nodes that can be arranged in any user defined order, and each node is evaluated to return a time array that may be used to populate a time series. A layering framework may be used to define how each node is evaluated.

Abstract: Disclosed are systems and methods providing for automation of enterprise and other processes. The systems and methods involve receiving historical process data, applying process mining techniques and generating process models. The process models can be used to identify automation candidates. One or more automation tools designed and configured for the identified automation candidates can be deployed to automate or to increase the efficiency of the process. In one embodiment, automation tools include artificial intelligence networks, which can label a set of input data according to determined or preconfigured domain-specific labels. An aggregator module can combine the similarly labeled data as part of automating a process or to increase the efficiency of a process.

Abstract: In some implementations, a method includes obtaining an unlabeled computer security data log and processing the unlabeled computer security data log using a machine learning model to generate a probability distribution that includes a respective probability for each of a plurality of possible log types. Each of the plurality of possible log types is associated with a corresponding parser that parses logs of the possible log type to extract structured computer security data. The method further includes selecting the possible log type having the highest probability and parsing the unlabeled computer security data log using the parser corresponding to the selected possible log type.