Abstract: Disclosed herein is a computer-implemented method for the ingestion of data into a partitioned database, the method comprising: receiving data at at least one ingestion node of a graph database, storing the data as a disjoint set of vertices, in a partitioned database, analyzing the disjoint set of vertices to find a set of remote edges and a set of native edges, and storing the set of remote edges and the set of native edges as a set of disjointed vertices in the partitioned database.

Abstract: Techniques for recommending hashtags, including trending hashtags, are disclosed. An example method includes accessing a graph. The graph includes video nodes representing videos, historical hashtag nodes representing historical hashtags, and edges indicating associations among the video nodes and the historical hashtag nodes. A trending hashtag is identified. An edge is added to the graph between a historical hashtag node representing a historical hashtag and a trending hashtag node representing the trending hashtag, based on a semantic similarity between the historical hashtag and the trending hashtag. A new video node representing a new video is added to the video nodes of the graph. A graph neural network (GNN) is applied to the graph, and the GNN predicts a new edge between the trending hashtag node and the new video node. The trending hashtag is recommended for the new video based on prediction of the new edge.

Abstract: A processor-implemented method for generating a schedule for executing operations of a compute graph includes receiving a graph including multiples nodes connected by edges. Each of the multiple nodes represents an operation to be executed. A set of sequences for executing the nodes is determined based on one or more precedence constraints. One or more sequences are selected from the set of sequences based on a memory constraint associated with a device for executing the nodes. A schedule for executing the nodes on the device is generated based on the selected one or more sequences.

Abstract: A database system generates an execution plan including multiple operators for processing a database query, for example, a join query or a group by query. The database system allocates a set of threads. Threads communicate with other threads via blocking channels. A blocking channel includes a buffer of a fixed capacity. The database system processes the database query by streaming data through operators of the execution plan. A thread sends data generated by an operator to another thread via the blocking channel if the buffer of the blocking channel has available capacity to store the data, or else the thread blocks until the buffer has capacity to store the data. Similarly, a thread receives data generated by an operator of another thread via the blocking channel if the buffer of the blocking channel has available data, or else the thread blocks until the buffer has data.

Abstract: Systems and methods are provided for enhanced processing of time series data via parallelization of instructions. An example method includes receiving a query indicating time series datasets and operations to be performed on the time series datasets. Nodes associated with the query are identified, with each node associated with a time series dataset. Nodes associated with operations to be performed are generated. The nodes are assembled into query tree, with parent nodes of the query tree indicating operations that are to be applied to children nodes. Instructions for processing the query tree are generated. At least a subset of the instructions is provided to one or more compute systems for processing in parallel. Results are received, and presented in a user interface.

Abstract: Systems, methods, and non-transitory computer-readable media can be configured to determine a set of candidates based at least in part on filtering criteria. A subset of candidates can be determined from the set of candidates based at least in part on one or more recruiter features associated with a recruiter. A recommendation can be provided to the recruiter for a candidate from among the subset of candidates based at least in part on a ranking of the subset of candidates.

Abstract: The present disclosure is directed toward systems, methods, and non-transitory computer readable media for generating and suggesting content collections for user accounts of a content management system using combinations of content-based features such as textual signals and visual signals. In some embodiments, the disclosed systems select a seed content item from among a plurality of content items associated with a user account within a content management system. From the seed content item, the disclosed systems can determine one or more germane topics and can cluster additional content items in relation to the germane topic(s). In addition, the disclosed systems can select one or more content items from a content cluster to provide as a suggested content collection.

Abstract: The disclosed method may include (i) aggregating, within a group network application, program user accounts registered to perform a respective task defined in a database of the group network application with resource user accounts registered as sources to provide resources for performance of the respective task, (ii) provisioning an interface that enables a program user account to select a structure for a metadata publication from a set of predefined structures, (iii) fulfilling at least part of the metadata publication based on an instruction from the program user account, and (iv) publishing the metadata publication over a wide area network in accordance with the instruction from the program user account.

Abstract: One example method includes associating a plurality of documents with a project, each document of the plurality of document having an associated status; generating a project status based on the status of each document of the plurality of documents; receiving an updated status of a first document; adjusting the project status based on the updated status; and in response to the updated project status meeting a threshold, providing a notification of the project status.

Abstract: Techniques are described herein that are capable of increasing security of a computer program using unstructured text. Unstructured text is received from web-based sources. The unstructured text includes user-generated posts. A machine learning model is trained by determining each keyword of a plurality of keywords in the unstructured text that corresponds to a computer program and further by determining each keyword of the plurality of keywords in the unstructured text that corresponds to a security vulnerability. The user-generated posts that are included in the unstructured text are filtered, using the machine learning model, to provide a subset of the user-generated posts such that each user-generated post in the subset includes a keyword that corresponds to the computer program and a keyword that corresponds to the security vulnerability. An action is performed based at least in part on the subset of the user-generated posts.

Abstract: The present application discloses a method, system, and computer system for automatically detecting protocol compliance of applications. The method includes determining a URL of a webpage for a software-as-a-service (SaaS) product, extracting body text from the webpage, and using a classifier to determine whether the SaaS product is compliant with one or more protocols.

Abstract: A portable information device includes: an image sensor used to pick up an image of a subject; a locator configured to acquire position information about an image pickup position for the subject; an orientation sensor configured to acquire an orientation information corresponding to an imaging direction at the image pickup position; a display configured to, while the subject is being imaged via the image sensor, display (a) an imaging data of the subject obtained via the image sensor and (b) related information on the subject obtained based on acquired data from the locator and from the orientation sensor; an input portion with which a user posts comment, while the subject is being imaged via the image sensor; and a controller configured to, while the subject is being imaged via the image sensor, control the display to display the posted comment and the related information.

Abstract: The disclosure is the only product of its kind that elevates the old passing of a paper note to a high-tech option that employs messaging technology to provide users with the convenience of chatting directly via on-line dating/meet-up sites that are disconnected from one's personal contact information, but with people they already have come across in person. This unprecedented device is uniquely designed to offer users with a new and improved means of connecting with others via poker chip that works in complement with a state of the art interactive website providing an alternative means to engage with new prospects, whether for social, personal or business needs.

Abstract: A method of searching and recommending online consumable items based on quantified benefit value, comprising receiving one or more benefit parameters defined by one or more users with respect to one or more domains of interest, searching for a plurality of consumable items relating to the interest domain(s) which are published in one or more online resources, analyzing a benefit label associated with each consumable items which is indicative of one or more quantified benefit values of the respective consumable item expressed in one or more metrics—time value, monetary value, and/or life improvement value, selecting one or more of the consumable items based on a correspondence between their quantified benefit values and the benefit parameter(s), and instructing one or more client devices used by the one or more users to present the selected consumable item(s) for consumption by the one or more users.

Abstract: Implementations of the present disclosure include actions of receiving search results that are responsive to the search query, the search results being respectively representative of resources, processing one or more triggering rules based on at least one of the search query and the search results, determining, based on processing the one or more triggering rules, that a set of data records is to be displayed in a search results page with the search results, the set of data records including data records that are specific to the user, determining a display prominence of at least one data record in the set of data records, the display prominence indicating a prominence of the at least one data record when displayed in the search results page, and providing the search results and the set of data records for display, the at least one data record being displayed based on the display prominence.

Abstract: Apparatuses, systems, methods, and computer program products are disclosed for an aggregated data query interface. A method includes authenticating a user of a publicly accessible query interface. A method includes authenticating the user for a third-party financial account of the user. A method includes accessing financial transaction data from the third-party financial account on behalf of the user. A method includes receiving a query from the user via the publicly accessible query interface. A method includes embedding, in a graphical user interface on an electronic display screen, a display of one or more financial transactions from the financial transaction data from the third-party financial account with a display of query results from the publicly accessible query interface, the one or more financial transactions and the query results each associated with the query from the user.

Abstract: A webpage display method, including: monitoring loading conditions of a display area of a home page, performing a node insertion measure to obtain a floating display area, and then executing a floating item configuration measure in the display area of the home page; monitoring a control action of a control device in the display area of the home page, and performing an information windowing measure to obtain floating webpages; and monitoring the number of floating webpages, monitoring the control action of the control device in the floating display area, and then performing an action feedback measure. The present method may obtain the floating display area by means of initialization in different browsers, feed back mouse click actions of a user on a floating page by means of different measures, and display content of any type by means of the floating page in a floating manner, thereby greatly increasing the efficiency with which a browser processes multiple tasks.

Abstract: An image generation system, comprising: a determination unit configured to determine whether or not a URL received from a communication apparatus is registered in a database; and a transmission unit configured to transmit, in a case where it has been determined that the URL is registered, a URL of a Web image, which Web contents corresponding to the URL have been converted into, to the communication apparatus.

Abstract: This application provides techniques for presenting multimedia resources. The techniques comprises obtaining a browsing instruction submitted by a user, and sending a browsing request to a server based on the browsing instruction; receiving a multimedia resource delivered by the server for the browsing request, and generating a browsing page based on the multimedia resource and displaying the browsing page to the user; in response to receiving a touch instruction performed by the user on the browsing page, determining multiple sub-resources with an association relationship from the multimedia resource based on the touch instruction; and generating a display page including at least two sub-resources based on the multiple sub-resources, and displaying the display page to the user.

Abstract: A method of aggregating displays of performances into an aggregate site on a network is provided. The aggregated performances originate from at least one performance site on a network. The method includes the steps of selecting a performance criterion; observing at least one performance originating from at least one performance site on a network, the performance being associated with a link; determining when at least one performance meets the performance criterion; establishing an aggregation link to the link associated with the performance meeting the performance criterion; and providing the aggregation link to an aggregate site on a network such that the performance is accessible on the aggregate site.

Abstract: Systems and methods are described that include generating a repository of metadata based on a plurality of webpages accessed and saved in a browser history of a web browser executing on a computing device, generating, based on the metadata, a history cluster including a portion of the plurality of webpages related to a topic where the history cluster generation is based on the source events and the access timestamps of the webpages in the portion, and assigning respective scores for the webpages in the portion. In response to a request to view browser activity associated with the topic, the systems and method may generate and display a history cluster listing for the topic where the history cluster listing includes visit listings associated with the webpages in the history cluster that are determined to have a score that meets a threshold score.

Abstract: A webpage scoring method used with a search engine and an electronic device are disclosed. The method includes: analyzing a webpage content of a target webpage to obtain environmental rating information corresponding to the target webpage; performing a keyword search by the search engine; and presenting a search result related to the target webpage with the environmental rating information in a window interface.

Abstract: A method of generating value based consumable items, comprising receiving one or more of a plurality of benefit parameters defined with respect to one or more of a plurality of interest domains selected by one or more users using respective client devices, generating one or more generated consumable items relating to the one or more interest domains using one or more generative Machine Learning (ML) models, computing a quantified benefit value for each of the plurality of generated consumable items, selecting one or more of the generated consumable items according to a correspondence between the quantified benefit value(s) of the respective generated consumable item and the one or more benefit parameters, and instructing one or more of client devices to present the selected one or more generated consumable items for consumption by one or more of the users.

Abstract: A system and method for analyzing in-page behavior. A method includes recording sessions of users browsing a website, wherein a session is time-ordered collection of a user's interactions with one or more webpages belonging to the website; analyzing recorded sessions to generate session insights, wherein the session insights are based in part on user experience factors, wherein each user experience factor relates to behavior of a user within each webpage visited during a session; and reporting the generated experience insights.

Abstract: The present disclosure relates to systems and methods for automatically identifying additional content for a webpage. The systems and methods may create a side pane placed next to a webpage to supplement the webpage in a browser. The side pane may provide additional information or additional content to help users consume the webpage. The additional content may provide webpage insights and help the users further explore the webpage.

Abstract: A system and method that converts the digital typesetting documents used in publishing to a device-specific format for electronic publishing. A “smart file and device-specific application” approach maintains the “look and feel” (design) of the source document used for print publication while typesetting for a specific device. Although this approach requires considerably more resources to create a smart file for each device-specific format, the smart file retains the unique typesetting characteristics of the printed book, is more aesthetically pleasing, and is easier to read. Furthermore, the device-specific application can render the smart file more quickly thereby eliminating any latency.

Abstract: The disclosed technology provides for generating a lossless representation of a webpage on a client device. A method can include rendering, by a client device, a webpage to be presented in a web browser at the client device, receiving an indication of user input to run a content check on the rendered webpage, calling, in response to receiving the indication of user input, a scraping function to scrape a document object model (“DOM”) of the rendered webpage in the web browser at the client device, generating a lossless representation of the scraped DOM of the rendered webpage, and transmitting, to a web content analysis system, the lossless representation of the scraped DOM to be used when running the content check for the rendered webpage. The lossless representation of the scraped DOM can be a flat object model (“FlatOM”) of the DOM of the rendered webpage.

Abstract: Techniques for providing a cross-ecosystem website for applications are presented. The techniques can include storing, by a webserver, multiple website templates in association with a plurality of respective web widgets and a respective Application Program Interface (API), where a plurality of respective back-end services correspond to the plurality of respective web widgets; receiving, from a website developer, a website template selection and a web widget selection; implementing a respective API for the cross-ecosystem website; storing in association with the cross-ecosystem website a manifest indicating services available on the cross-ecosystem website; and interfacing with a plurality of applications in a plurality of application ecosystems, using the API for the cross-ecosystem website.

Abstract: Provided herein are systems, methods and devices for classifying nested content execution loaded by a webpage or an application executed by a client device, comprising a client device executing a webpage or an application loaded from a content server which embed nesting element(s) used for loading nested content from nested content server(s). The webpage/application embeds a host monitoring code executed to collect session data indicative of execution session of the webpage/application including execution of nested content loaded using the nesting element(s), transmitting a signature of the session data to server(s) configured to classify the execution according to at least part of the session data extracted from the signature, and transmitting transaction indicator(s) of execution of the nested content to one or more providers of the nested content which may verify execution of the nested content in the context of the webpage/application based on the classification obtained from the server(s).

Abstract: In an optimization device, an acquisition means acquires a reward obtained by executing a certain policy. An updating means updates a probability distribution of the policy based on the obtained reward. Here, the updating means uses a weighted sum of the probability distributions updated in a past as a constraint. A determination means determines the policy to be executed, based on the updated probability distributions.

Abstract: A compute-in-memory (CIM) circuit includes a processing circuit. The processing circuit includes a data-selection circuit and a charge-domain passive summation circuit. The data-selection circuit includes a memory array and a selection circuit. The memory array stores a plurality of candidate weights. The selection circuit selects a target weight from the plurality of candidate weights stored in the memory array. The charge-domain passive summation circuit generates an analog computation result of an input received by the processing circuit and the target weight stored in the memory array through a weighted capacitor array integrated with the memory array.

Abstract: A data processing method and data processing apparatus are provided. The data processing method includes: acquiring multiple input tensors as input parameters for calculation process; for each input tensor, using M input sub-tensors that are combined to represent the input tensor; for each of the input tensors, replacing the input tensors with the M input sub-tensors that are combined to represent the input tensor, and performing the calculation process to obtain a calculation result. The data processing method increases the applicable scenarios of calculation process, effectively utilizes the powerful calculation ability of the originally provided low-accuracy floating points, and greatly improves the overall calculation efficiency.

Abstract: In examples, a device comprises control logic configured to detect an idle cycle, an operand generator configured to provide a synthetic operand responsive to the detection of the idle cycle, and a computational circuit. The computational circuit is configured to, during the idle cycle, perform a first computation on the synthetic operand. The computational circuit is configured to, during an active cycle, perform a second computation on an architectural operand.

Abstract: In a method a first and a second column-split matrix comprise columns of a left side matrix and a first and a second row-split matrix comprise rows of a right side matrix. A Matrix Processing Unit (MPU) receives column elements of a row of the first column-split matrix and row elements of a column of the first column-split matrix, A second MPU receives column elements of a row of the second column-split matrix and row elements of a column of the second row-split matrix. The first and second MPU concurrently compute partial dot products of the column and row elements and a third MPU computes a sum of the partial dot products. A computing system can include the MPUs and can implement the method.

Abstract: In a method, based on a left side matrix and a right side matrix having a shared dimension, a first Multiply Accumulate Arithmetic Logic Unit (MACC ALU) receives elements of a row of a first column-split matrix and elements of a column of a first column-split matrix. A second MACC ALU receives elements of a row of the second column-split matrix and elements of a column of the second row-split matrix. The first and a second column-split matrices comprise columns of the left side matrix and the first and second row-split matrices comprise rows of the right side matrix. The first and second MACC ALU concurrently compute partial dot products of the column and row elements and the second MACC ALU computes a sum of the partial dot products. A computing system can include the MACC ALUs in a matrix processing unit and can implement the method.

Abstract: The present disclosure relates to methods and apparatus for compute processing. For example, disclosed techniques facilitate improving performance of matrix multiplication in streaming processor. Aspects of the present disclosure can execute, with a load control unit, a first load instruction to load a set of input data of an input matrix from a first memory to a second memory. Aspects of the present disclosure can also execute, with the load control unit, a second load instruction to load a set of weight data of a weight matrix from the first memory to the second memory. Additionally, aspects of the present disclosure can perform, with an ALU component, a matrix multiplication operation using the set of input data and the set of weight data to generate an output matrix. Further, aspects of the present disclosure can store the output matrix at a general purpose register accessible to the ALU component.

Abstract: A sampling system includes: an energy-based model (EBM) configured to generate non-negative scores of an input having discrete classifications, respectively; and a sampling module configured to: generate a sample from a probability distribution of the EBM using a proposal distribution; set a probability of acceptance of the sample based on a minimum of (a) 1 and (b) an acceptance value determined based on the sample, a score of the sample from the EBM, the proposal distribution, and an upper boundary value; determine a distribution value between 0 and 1 using a uniform distribution; and discard the sample when the distribution value is greater than the probability of acceptance of the sample.

Abstract: The inconsistency determination device 1 includes inconsistency determination means 9. The inconsistency determination means 9 determines whether or not there is inconsistency in an expression, which is to be converted into an energy function of a model representing states of individual spins by a first value or a second value, representing energy in a combinatorial optimization problem.

Abstract: Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.

Abstract: Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.

Abstract: Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.

Abstract: Methods, systems, and non-transitory computer-readable media are configured to perform operations comprising determining at least one criterion for generation of augmented data to be included in a set of training data for training a machine learning model. At least one base template and at least one component are selected based on the at least one criterion. The augmented data is generated based on the at least one base template and the at least one component.

Abstract: An item classification method and system using multi-output headed ensembles, that can include receiving one or more text input sequences at one or more first estimator threads corresponding to the one or more text input sequences. The method can also include tokenizing the one or more text input sequences into one or more first tokens within the one or more first estimator threads. In addition, the method can include outputting one or more item classifications based on an output of the one or more first estimator threads. Further, the method may include applying a backpropagation algorithm to update network weights connecting neural layers in the first estimator threads, defining an optimal setting of network parameters using cross-validation with respect to the first estimator threads, and mapping the one or more first tokens to an embedding space within the one or more first estimator threads.

Abstract: In recommender systems, the goal is to learn a model from a small set of interacted users and items, and identify positively-related user item pairs among a large number of pairs with unknown interactions. Known methods for training the model rely on both similar and dissimilar user-item pairs. Using dissimilar pairs introduces several challenges, such as increasing training time or labeling pairs with unknown interactions as dissimilar even though the user might like the item if presented with it. If only similar pairs are used in the known methods, the result is a collapsed solution in which all users and items are mapped to the same representations. The methods disclosure herein overcome these challenges by using only similar pairs but adding two terms to the objective function that prevent a collapsed or partially-collapsed solution.

Abstract: Secure Real Time Communications Service (SRTCS) for audio and video streaming communications and content sharing that securely connects multiple users using a “push-button” WebRTC chat app connection over a Peer-to-Peer (P2P) network. SRTCS uniquely combines advanced security technologies to provide user based permissions control when communicating and sharing rich media content with other users including End-to-End Encryption (E2EE), Hash Technology (DHT), and Digital Rights Protection (DRM). SRTCS has also designed a unique cloud based streamed video storage and sharing platform service for consumers and business video storage and sharing applications.

Abstract: The current document is directed to reinforcement-learning-based management-system agents that control distributed applications and the infrastructure environments in which they run. Management-system agents are initially trained in simulated environments and specialized training environments before being deployed to live, target distributed computer systems where they operate in a controller mode in which they do not explore the control-state space or attempt to learn better policies and value functions, but instead produce traces that are collected and stored for subsequent use. Each deployed management-system agent is associated with a twin training agent that uses the collected traces produced by the deployed management-system agent for optimizing its policy and value functions. When the optimized policy is determined to be more robust, stable, and effective than the policy of the corresponding deployed management-system agent, the optimized policy is transferred to the deployed management-system agent.

Abstract: Embodiments disclosed are directed to a computing system that performs steps for third-party account creation with an authentication implementation that authenticates a third-party user while avoiding the need for the third-party user to create a password-protected account. The computing system generates a service request associated with an entity and a digital item of record stored in a records database. The computing system then generates a first unique key associated with the entity and the digital item of record and embeds the first unique key as a first clickable link in the service request. Subsequently, in response to receiving an electronic indication that the entity has clicked the first clickable link, the computing system verifies that an identity of the entity is authentic based on the first unique key.

Abstract: Arrangements for providing frictionless unauthorized activity detection and user authentication are provided. In some aspects, user data, such as transaction data may be received and used to train a machine learning model. In some examples, the machine learning model may be executed to generate one or more expected user patterns. In some arrangements, a request for transaction may be received. The request for transaction may include transaction details. In response, the system may request current geo-location data of a user. In some examples, the transaction details and geo-location data may be analyzed (e.g., compared to the expected user patterns) to generate an authentication output. The authentication output may then be transmitted to one or more systems to process the requested transaction, prevent transaction processing, or the like.

Abstract: Embodiments of the present invention provide a system for interconnection, translation, and transition between disparate digital ecosystems. The system is configured for determining that a user is requesting to access a first digital ecosystem, receiving a first authentication credentials of the user from the first digital ecosystem for verification, determining that the verification of the first authentication credentials is successful and provide access to the first digital ecosystem, determining that the user is requesting transition to a second digital ecosystem from the first digital ecosystem, receiving second authentication credentials of the user from the second digital ecosystem for verification, determining that the verification of the second authentication credentials is successful and provide access to the second digital ecosystem, and preparing and transmitting a data payload to the second digital ecosystem.

Abstract: Systems, computer program products, and methods are described herein for identification and verification of brute force attacks using hash signature segmentation and behavior clustering analysis. The present invention is configured to receive a plurality of access attempts and an associated plurality of hash values; determine a failure request load based on the plurality of access attempts; determine whether the failure request load meets a failure threshold, wherein, in an instance where the failure request load meets the failure threshold, separating the plurality of access attempts associated with the failure request load into a plurality of hash clusters; generate a behavior cluster for each hash cluster based on shared behavior data of the plurality of access attempts; and determine, based on the behavior cluster for each hash cluster, a likelihood of misappropriation for each access attempt of the plurality of access attempts.