Abstract: A method of analyzing a performance of a microservices-based application comprises generating a plurality of traces from a plurality of spans associated with the microservices-based application. The method also comprises generating a plurality of data sets each associated with a respective analysis mode of a plurality of analysis modes using the plurality of traces, wherein each analysis mode extracts a different level of detail for analyzing the performance of the services in the application from the plurality of spans. Further, the method comprises selecting, based on a first user query, a first analysis mode from the plurality of analysis modes for generating a response to the first user query. The method also comprises accessing a data set of the plurality of data sets that is associated with the first analysis mode and generating the response to the first user query using the data set associated with the first analysis mode.

Abstract: One or more unit-test cases are generated from a monolingual code corpus and the generated unit-test cases are filtered to generate a corpus of unit-test cases which have acceptability scores exceeding one or more predefined thresholds. One or more of the code samples of the monolingual code corpus are translated from a source language to a target language using a pretrained Large Language Model and the generated unit-test cases are translated from the source language to the target language. The LLM-translated code samples are validated using the translated unit-test cases and a parallel-data training corpus comprising the LLM-translated code samples that pass the validation is created. The pretrained large language model (LLM) is fine-tuned using the parallel-data training corpus, a given code segment is translated using the fine-tuned large language model (LLM), the translated given code segment is tested and the tested given code segment is deployed.

Abstract: An entity-level visibility statistic may be determined for a database entity in a database system based on one or more visibility rules providing access to instances of the database entity to one or more user accounts. A user-level visibility statistic quantifying a set of instances of the database entity accessible to a user account via the one or more visibility rules may be determined based at least in part on the entity-level visibility statistic. A request may be received by the user account to execute an input database query retrieving one or more of the instances of the database entity. A database object retrieval query including two or more data security subqueries evaluating accessibility of the one or more instances of the database entity and positioned based at least in part on the user-level visibility statistic may be determined based on the input database query.

Abstract: The innovations disclosed herein concern improvements for a hosted gaming environment to provide improved flexibility, scalability, and game deployment/maintenance. In particular, the multi-form gaming platforms disclosed herein may host so-called “iGaming” sessions (i.e., online gaming with real money wagering), mobile Class-II gaming sessions (e.g., Bingo games) and/or mobile Class-III gaming sessions (e.g., traditional casino-style games), as well as “Social” or “Demo” gaming sessions (i.e., online gaming with non-real money wagering).

Abstract: A novel formulation called the Context-Attentive Bandit with Observations (CABO) is described, where only a limited number of features can be accessed by the learner. The present invention is applicable to many problems including problems arising in clinical settings and dialog systems where it is not possible to reveal the whole feature set. The present invention adapts the standard contextual bandit algorithm known as Thompson Sampling with a novel algorithm, we call Context-Attentive Thompson Sampling with Observations (CATSO). Experimental results are included to demonstrate its effectiveness including a regret analysis and an empirical evaluation demonstrating advantages of the disclosed novel approach on several real-life datasets.

Abstract: In order to extract more contextualized metrics associated with a particular service in a microservices-based application, a method of aggregating metrics comprises analyzing tags of spans for each trace and, based on a first predetermined set of rules, extracting one or more tags pre-selected by a user of the monitoring service. A pre-selected tag may appear in one or more spans associated with a particular service in a trace and may be extracted as a numeric metric. The method comprises computing a set of metrics for the particular service using the numeric metric extracted from the one or more spans of the trace. The method also comprises determining a set of aggregated metrics over the selected time duration for the particular service by aggregating the set of metrics over multiple traces.

Abstract: An entity-level visibility statistic may be determined for a database entity in a database system based on one or more visibility rules providing access to instances of the database entity to one or more user accounts. A user-level visibility statistic quantifying a set of instances of the database entity accessible to a user account via the one or more visibility rules may be determined based at least in part on the entity-level visibility statistic. A request may be received by the user account to execute an input database query retrieving one or more of the instances of the database entity. A database object retrieval query including two or more data security subqueries evaluating accessibility of the one or more instances of the database entity and positioned based at least in part on the user-level visibility statistic may be determined based on the input database query.

Abstract: A method of rendering a service graph illustrating dependencies between a frontend and a backend of an application comprises generating a plurality of frontend traces from a plurality of frontend spans and generating a plurality of backend traces from a plurality of backend spans ingested from the application. The method also comprises aggregating frontend metrics data using the plurality of frontend traces and backend metrics data using the plurality of backend traces. The method further comprises determining connection information between one or more frontend traces of the plurality of frontend traces and corresponding backend traces of the plurality of backend traces. The method also comprises rendering the service graph using the connection information and the aggregated frontend and backend metrics data.

Abstract: A method of aggregating metrics associated with a user interaction during a real user session comprises identifying a span comprising a tag associated with a workflow from ingested spans associated with the real user session, where the workflow comprises spans generated in response to the user interaction. The method also comprises identifying other spans associated with the workflow from the ingested spans. The method further comprises grouping the other spans associated with the workflow with the tagged span and aggregating metrics for the workflow over a duration of time.

Abstract: A method of rendering a service graph illustrating dependencies between a frontend and a backend of an application comprises generating a plurality of frontend traces from a plurality of frontend spans and generating a plurality of backend traces from a plurality of backend spans ingested from the application. The method also comprises aggregating frontend metrics data using the plurality of frontend traces and backend metrics data using the plurality of backend traces. The method further comprises determining connection information between one or more frontend traces of the plurality of frontend traces and corresponding backend traces of the plurality of backend traces. The method also comprises rendering the service graph using the connection information and the aggregated frontend and backend metrics data.

Abstract: One or more unit-test cases are generated from a monolingual code corpus and the generated unit-test cases are filtered to generate a corpus of unit-test cases which have acceptability scores exceeding one or more predefined thresholds. One or more of the code samples of the monolingual code corpus are translated from a source language to a target language using a pretrained Large Language Model and the generated unit-test cases are translated from the source language to the target language. The LLM-translated code samples are validated using the translated unit-test cases and a parallel-data training corpus comprising the LLM-translated code samples that pass the validation is created. The pretrained large language model (LLM) is fine-tuned using the parallel-data training corpus, a given code segment is translated using the fine-tuned large language model (LLM), the translated given code segment is tested and the tested given code segment is deployed.

Abstract: A method of persisting and querying Real User Monitoring (RUM) data comprises grouping together spans associated with a user-interaction with a webpage or application that are ingested during a given time duration. The method also comprises generating one or more data sets each associated with an analysis modality using the grouped spans, wherein each analysis modality extracts a different level of detail from the spans. Further, the method comprises selecting, based on a first user query, a first analysis modality for generating a response to the first user query and accessing a data set that is associated with the first analysis modality. The method also comprises generating the response to the first user query using the data set associated with the first analysis modality, wherein the first user query requests information pertaining to a performance of the application in response to the user-interaction.

Abstract: The innovations disclosed herein concern improvements for a hosted gaming environment to provide improved flexibility, scalability, and game deployment/maintenance. In particular, the multi-form gaming platforms disclosed herein may host so-called “iGaming” sessions (i.e., online gaming with real money wagering), mobile Class-II gaming sessions (e.g., Bingo games) and/or mobile Class-III gaming sessions (e.g., traditional casino-style games), as well as “Social” or “Demo” gaming sessions (i.e., online gaming with non-real money wagering).

Abstract: A method of tracking errors in a system comprising microservices comprises ingesting a plurality of spans generated by the microservices during a given duration of time. The method further comprises consolidating the plurality of spans associated with the given duration of time into a plurality of traces, wherein each trace comprises a subset of the plurality of spans that comprise a common trace identifier. For each trace, the method comprises: a) mapping a respective trace to one or more error stacks computed for the respective trace and to one or more attributes determined for the respective trace; and b) emitting each error stack computed from the respective trace with an associated pair of attributes. The method then comprises reducing duplicate pairs of error stack and associated attributes and maintaining a count for each pair of error stack and associated attributes.

Abstract: A mobile support system for providing wagering game play to a plurality of mobile devices is provided. The system is configured to receive a play request for the wagering game from a mobile device, where the play request includes a wager amount and represents placement of a wager for an instance of game play of the wagering game. The system generates an instance of game source data for the wagering game and identifies a game play data set used for the play request. Prior to performing a withdrawal transaction of the wager amount for the play request, the system generate a wager outcome for the play request by comparing the game source data with the game play data set. After determining the wager outcome for the play request, the system causes the withdrawal transaction of the wager amount to be performed, thereby completing a placement of the wager amount.

Abstract: Monitoring and troubleshooting tools provide the capability to visualize different levels of a client's application that is deployed as a suite of independent but cooperating services (e.g., an application that includes a monolithic application and a microservices-based application), collect values of monitored or tracked metrics at those different levels, and visualize values of the metrics at those levels. For example, metrics values can be generated for components of the monolithic application and/or for components of a microservice of the microservice-based application.

Abstract: A method of aggregating metrics associated with a user interaction during a real user session comprises identifying a span comprising a tag associated with a workflow from ingested spans associated with the real user session, where the workflow comprises spans generated in response to the user interaction. The method also comprises identifying other spans associated with the workflow from the ingested spans. The method further comprises grouping the other spans associated with the workflow with the tagged span and aggregating metrics for the workflow over a duration of time.

Abstract: A method, computer program, and computer system is provided for testing a user interface. A previously trained machine learning model trained with traces of interactions between one or more users and a user interface is accessed. The interactions include one or more timestamps of user interactions with the user interface, actions by each user associated with the user interface, and metadata associated with user interactions. A simulated interaction of a simulated agent utilizing the user interface is generated using the previously trained machine learning model. The simulated interaction is encoded as an input trace to a user interface. The encoded simulated interaction is input into the user interface for automated testing of the user interface. Results of the automated testing of the user interface are received.

Abstract: In federated learning problems, data is scattered across different servers and exchanging or pooling it is often impractical or prohibited. A Bayesian nonparametric framework is presented for federated learning with neural networks. Each data server is assumed to provide local neural network weights, which are modeled through our framework. An inference approach is presented that allows us to synthesize a more expressive global network without additional supervision, data pooling and with as few as a single communication round. The efficacy of the present invention on federated learning problems simulated from two popular image classification datasets is shown.

Abstract: A method of analyzing a performance of a microservices-based application comprises generating a plurality of traces from a plurality of spans associated with the microservices-based application. The method also comprises generating a plurality of data sets each associated with a respective analysis mode of a plurality of analysis modes using the plurality of traces, wherein each analysis mode extracts a different level of detail for analyzing the performance of the services in the application from the plurality of spans. Further, the method comprises selecting, based on a first user query, a first analysis mode from the plurality of analysis modes for generating a response to the first user query. The method also comprises accessing a data set of the plurality of data sets that is associated with the first analysis mode and generating the response to the first user query using the data set associated with the first analysis mode.