Abstract: In various example embodiments, a system and method for adjusting a user interface according to a cognitive mode is disclosed. In one example embodiment, a method includes receiving a plurality of user events based on a user interacting with a system, determining a cognitive operating mode of the user based on the user events, with the cognitive operating mode selected from a predefined set of cognitive operating modes, and modifying a user interface element according to the cognitive operating mode of the user.

Abstract: Summaries of media programs that are in progress are generated based on content of the media programs that has already been transmitted to listeners or viewers. The content is transcribed into text, and contextual features regarding the media program such as topics, identities of speakers or interactions received from listeners are identified. The transcribed content and the contextual features are provided as multi-modal inputs to a model that is trained to generate a summary of the media program in response to such inputs. Summaries of media programs that are then in progress are transmitted to devices of listeners who may be interested in joining one of the media programs and displayed in a menu or user interface or announced to the listeners.

Abstract: The present disclosure generally relates to interfaces and techniques for media playback on one or more devices. In accordance with some embodiments, an electronic device includes a display, one or more processors, and memory. The electronic device receives user input and, in response to receiving the user input, displays, on the display, a multi-device interface that includes: one or more indicators associated with a plurality of available playback devices that are connected to the device and available to initiate playback of media from the device, and a media playback status of the plurality of available playback devices.

Abstract: A computer device is configured to provide a computer implemented game. At least one processor of the computer device is configured to determine that one more matches have been made by a plurality of game elements in a game board area and in response to provide one or more game entities in the game board area. The at least one processor is configured to control one or more of the game entities to move from the game board area to an interaction area.

Abstract: A computer device has a user interface which receives user input to control movement of a game entity in a collection run of a computer implemented game. The computer device has a display which displays the game entity moving in response to the user input in the collection run. At least one processor of the computer device determines resources collected by the movement of game entity in the collection run. When one or more required resources have been collected, the at least one processor causes, in response to user input. one or more further items for the computer implemented game to be provided.

Abstract: A system is provided for tracing end-to-end transactions. The system uses bytecode instrumentation and a dynamically injected agent to gather web server side tracing data, and a browser agent which is injected into browser content to instrument browser content and to capture tracing data about browser side activities. Requests sent during monitored browser activities are tagged with correlation data. On the web server side, this correlation information is transferred to tracing data that describes handling of the request. This tracing data is sent to an analysis server which creates tracing information which describes the server side execution of the transaction and which is tagged with the correlation data allowing the identification of the causing browser side activity.

Abstract: According to one embodiment, a method, computer system, and computer program product for dynamically modifying interactive digital content based on viewer sentiment is provided. The present invention may include measuring, by one or more sensors, characteristics of viewers associated with a viewed portion of the interactive digital content, while the viewers are viewing the interactive digital content; inferring the sentiment of the viewers with respect to the viewed portion of the interactive digital content based on the measured characteristics; predicting content satisfying to the viewers based on the sentiment; and modifying the interactive digital content in real time based on the predicted content.

Abstract: One embodiment provides a method for federated learning across a plurality of data parties, comprising assigning each data party with a corresponding namespace in an object store, assigning a shared namespace in the object store, and triggering a round of federated learning by issuing a customized learning request to at least one data party. Each customized learning request issued to a data party triggers the data party to locally train a model based on training data owned by the data party and model parameters stored in the shared namespace, and upload a local model resulting from the local training to a corresponding namespace in the object store the data party is assigned with. The method further comprises retrieving, from the object store, local models uploaded to the object store during the round of federated learning, and aggregating the local models to obtain a shared model.

Abstract: A computer device having a user interface, a processor, and computer storage. The user interface provides a gameboard comprising tiles supporting game objects, the tiles comprising predefined sets of tiles which are available to support spreading blocking elements. The processor receives a detected user input. If it detects a match condition, the processor removes the game elements of the match condition and generates and displays new game objects to replenish the gameboard. A data structure having tile data is stored in the computer storage. Tile data for tiles of the predefined set(s) comprises a set indicator which indicates whether a tile is in one of the predefined sets. If a blocker spreading mode has been triggered, the processor selects an origin tile which supports a spreading blocking element, identifies a target tile to which the spreading blocking element may spread and generates a blocking element for the target tile.

Abstract: Systems and methods for managing food processing and inventory. Food inventory held at food holding locations can be monitored. Food can be transferred among holding locations and served from holding locations. A prioritized food type list indicates food types to be processed and lists food types in order of determined priority. Food may be transferred to default destination holding locations. Food may be moved to and served from primary holding locations. Buttons associated with holding location displays can be actuated by different types of user engagement for executing various food management operations.

Abstract: A method designs nuclear reactors using design variables and metric variables. A user specifies ranges for the design variables and threshold values for the metric variables and selects design parameter samples. For each sample, the method runs three processes, which compute metric variables for thermal-hydraulics, neutronics, and stress. The method applies a cost function to compute an aggregate residual of the metric variables compared to the threshold values. The method deploys optimization methods, either training a machine learning model using the samples and computed aggregate residuals, or using genetic algorithms, simulated annealing, or differential evolution. When using Bayesian optimization, the method shrinks the range for each design variable according to correlation between the respective design variable and estimated residuals using the machine learning model. These steps are repeated until a sample having a smallest residual is unchanged for multiple iterations.

Abstract: A computer-implemented method comprises receiving by a servers of a development platform, an upload of a trained ML model and configuration data from a tenant user, for use by a mobile enterprise application that is made available to one or more tenants of the development platform, the mobile enterprise application capable of invoking the trained ML model, wherein the configuration data defines an input data format and an output data format of the trained ML model. The server generates API based on the configuration data and downloading the API to the mobile enterprise application, wherein the generating the API is performed without using an API provided by the tenant user. The generated API is further configured to pass input in the input data format to the trained ML model and configured to receive output in the output data format from the trained ML model for further processing or display.

Abstract: A parser is deployed early in a machine learning pipeline to read raw data and collect useful statistics about the raw data's content to determine which items of raw data exhibit a proxy for feature importance for the machine learning model. The parser operates at high speeds that approach the disk's absolute throughput while utilizing a small memory footprint. Utilization of the parser enables the machine learning pipeline to receive a fraction of the total raw data that would otherwise be available. Several scans through the data are performed, by which proxies for feature importance are indicated and irrelevant features may be discarded and thereby not forwarded to the machine learning pipeline. This reduces the amount of memory and other hardware resources used at the server and also expedites the machine learning process.

Abstract: A system and method is disclosed for labeling an unlabeled dataset, with a labeling budget constraint and noisy oracles (i.e. noisy labels provided by annotator), using a noisy labeled dataset from another domain or application. The system and method combine active learning with noisy labels and active learning with domain adaptation to enhance classification performance.

Abstract: Embodiments described herein are directed to generating a machine learning (ML) model. A plurality of vectors are accessed, each vector of the plurality of vectors including a first set of features associated with a corresponding data item. A second set of features is identified by expanding the first set of features. A ML model is trained using vectors including the expanded set of features, and it is determined that an accuracy of the ML model trained using the vectors increased. A third set of features is identified by determining a measure of importance for different subsets of features in the second set and replacing subsets having a low measure of importance with new features. A ML model is trained using vectors that include the third set, and it is determined that an accuracy of the model increased due to the replacing.

Abstract: A question-and-answer application with an “ask-to-answer” feature is described. The ask-to-answer feature enables any user to solicit an answer to a question from another user. Upon soliciting another user for an answer to a particular question, a message with a call to action is directed to the solicited user. The message may include a copy of the text of the question and may provide a mechanism (e.g., a selectable user interface element) enabling the solicited user to pass on answering the question. Subsequent to the solicitation, the question page for the question will include a notification with information about the solicitation, including in some instances information identifying the user who has been asked to answer the question and the number of times the user has been asked to provide an answer to the question.

Abstract: Systems, methods, and non-transitory computer-readable media can be configured to acquire at least one instance of positive training data based at least in part on at least one source. A set of supplemental positive training data can be generated based at least in part on the at least one instance of positive training data. A machine learning model can be trained to identify inappropriate material based at least in part on the set of supplemental positive training data.

Abstract: A method for protecting a first machine learning (ML) model is provided. In the method, a dataset of non-problem domain (NPD) data is selected from a large dataset using a second ML model. The second ML model classifies the large dataset into NPD classifications and PD classifications. The PD classified data is excluded. A distinguisher includes a third ML model that is trained using selected NPD data from the large dataset. The distinguisher receives input samples that are intended for the first ML model. The third ML model provides either a PD classification or NPD classification in response to receiving each input sample. An indication of a likely extraction attempt may be provided when a predetermined number of NPD classifications are provided. The method provides an efficient way to create a training dataset for a distinguisher and for protecting a ML model with the distinguisher.

Abstract: Aspects described herein may provide systems, methods, and device for facilitating language learning using videos. Subtitles may be displayed in a first, target language or a second, native language during display of the video. On a pause event, both the target language subtitle and the native language subtitle may be displayed simultaneously to facilitate understanding. While paused, a user may select an option to be provided with additional contextual information indicating usage and context associated with one or more words of the target language subtitle. The user may navigate through previous and next subtitles with additional contextual information while the video is paused. Other aspects may allow users to create auto-continuous video loops of definable duration, and may allow users to generate video segments by searching an entire database of subtitle text, and may allow users create, save, share, and search video loops.

Abstract: A data processing method includes: obtaining a first subset and at least a second subset in a training dataset for learning, the first subset and the at least a second subset having a same size; determining a set of substrings based on data strings in the first subset and the at least a second subset, the substrings being suffix substrings of the data strings and being sorted in a lexicographical order; and determining a grain for deduplication of the training dataset from a set of longest common prefix (CLP) lengths of adjacent substrings in the set of substrings, for use in the deduplication. Thereby, different grains of duplicating procedures for different training datasets can be predicted automatically, and universality and flexibility of GPUaaS can be achieved. In addition, the deduplication rate can be improved, network resource waste can be reduced and system efficiency can be enhanced.