Abstract: The present disclosure discloses a reinforcement learning agent training method, modal bandwidth resource scheduling method and apparatus. The reinforcement learning agent training method utilizes a reinforcement learning agent to continuously interact with a network environment in a polymorphic smart network to obtain the latest global network characteristics and output updated actions. By adjusting the bandwidth occupied by modals, a reward value is set to determine an optimization target for the agent, the scheduling of modals is realized, and the rational use of polymorphic smart network resources is guaranteed. The trained reinforcement learning agent is applied to the modal bandwidth resource scheduling method, and can adapt to networks with different characteristics, and thus can be used for intelligent management and control of polymorphic smart networks and has good adaptability and scheduling performance.

Abstract: A method on a device in a content delivery (CD) network (CDN) that distributes content on behalf of one or more subscribers. In response to receiving configuration information from a subscriber, the configuration information relating to at least one property of the subscriber, generating subscriber-specific platform configuration information for the at least one property. Storing the subscriber-specific platform configuration information in platform configuration storage. Invalidating prior platform configuration information associated with the particular subscriber. Responsive to a request from a CDN component for platform configuration information associated with the particular subscriber: obtaining the subscriber-specific platform configuration information from the platform configuration storage; and providing the subscriber-specific platform configuration information to the CDN component.

Abstract: The present disclosure relates to for managing the generation or deletion of record objects based on electronic activities and communication policies. Data of a plurality of electronic activities from one or more electronic accounts of the entity may be identified. It may be determined whether the identified data satisfies a communication policy. The communication policy may include one or more rules based on the electronic activities. Instructions to generate a contact record object or instructions to delete an existing contact record for the entity may be transmitted based on the condition of the communication of the communication policy that is satisfied.

Abstract: A system can receive network build sheets corresponding to proposed network architectures for a computer network. The system can aggregate the network build sheets to generate the holistic build data for the computer network. The system can receive a request to access a user interface for displaying the holistic build data. The system can determine an access level associated with the request. The system can generate the user interface that includes a graphical representation of at least a portion of the holistic build data. The graphical representation can be customized based on the determined access level.

Abstract: An active edge caching method based on community discovery and weighted federated learning for selecting an optimal caching user and developing an optimal caching strategy; user grouping method based on community discovery is first provided, users are divided into different user groups according to users' mobility and social attributes, degrees of importance of different users are computed in each user group, and most important user is selected as a caching node to provide content distribution services; a content popularity prediction framework based on attention weighted federated learning is provided for combining attention weighted federated learning mechanism with deep learning algorithm to predict future user preferences for different content, so as to improve content popularity prediction accuracy and solve user privacy disclosure; finally, hot content is cached on selected caching users according to prediction results, to achieve goals of reducing network transmission latency and network operation cost

Abstract: Systems and methods for reducing wasted computational resources by detecting network operation validation anomalies in conglomerate-application-based ecosystems are disclosed. The system provides a first network operation to a first software application of a set of software applications, where (i) the first network operation is configured to be provided as input to the first software application and (ii) the first software application generates a second network operation as output, using the first network operation, to be provided to a second software application. The system receives a first message associated with a processing of the first network operation at the first software application. The system then performs a determined validation process on the first message. In response to detecting that the determined validation process results in a failure to validate the first message, the system prevents the second network operation from being provided to the second software application.

Abstract: Embodiments are directed to security analysis agents. Events associated with a computing environment may be provided. Prompt fragments may be determined based on the events. A prompt may be generated for a large language model (LLM) based on a prompt template and the prompt fragments such that the prompt fragments may be included in the prompt and provided to the LLM. Actions for evaluating the events may be determined based on the LLM response. These actions may be executed to evaluate the events. Portions of the response that correspond to the prompt fragments may be determined. A performance score may be determined for each prompt fragment based on its corresponding portion of the response such that the prompt may be modified to exclude a portion of the prompt fragments that have a performance score less than a threshold value.

Abstract: Systems and methods include receiving information from a governed system having resources; implementing one or more context neutral or specific control applications that each is a pipeline of a plurality of functions to control the resources; utilizing the information as one input to a first function in the plurality of functions; utilizing inputs from one or more control applications as inputs to any of the plurality of functions; and causing actions on the resources based on outputs of a last function in the plurality of functions. The plurality of functions can include sensing, discerning, inferring, deciding, and causing the actions.

Abstract: A system is provided for an artificial intelligence-based engine for generating recommendations for resource allocation. In particular, the system may comprise an artificial intelligence module configured to access one or more data sources containing resource data associated with one or more users. Based on the resource data, the system may identify a group of analogous users with respect to a specific user. The system may then generate one or more recommendations for a resource allocation associated with the user. The one or more recommendations may then be presented on a graphical interface of a user device associated with the user. In this way, the system may provide an AI-driven method for efficient allocation of resources.

Abstract: Cognitive functions in a network, each comprising a learning module adapted to observe characteristics of the network and an inference module adapted to conclude, from the characteristics observed by the learning module, on a behavior of the network resulting from the observed characteristics, are controlled by halting and/or restarting operation of the learning module of a cognitive function of the cognitive functions, where operation of the inference module of the cognitive function is controlled independently from the halting and/or restarting the operation of the learning module of the cognitive function.

Abstract: A process can include receiving monitoring information associated with a machine learning (ML) or artificial intelligence (AI) workload implemented by an edge compute unit of a plurality of edge compute units. Status information corresponding to a plurality of connected edge assets can be received, the plurality of edge compute units and connected edge assets included in a fleet of edge devices. A remote fleet management graphical user interface (GUI) can display a portion of the monitoring or status information for a subset of the fleet of edge devices, based on a user selection input, and can receive a user configuration input indicative of an updated configuration for at least one workload corresponding to a pre-trained ML or AI model deployed on the at least one edge compute unit. A cloud computing environment can transmit control information corresponding to the updated configuration to the at least one edge compute unit.

Abstract: A system and method for mapping and transforming communication stream messages to email messages are described. A communication stream message is communicated to a mapping engine which extracts elements of the communication stream message, maps and transforms the extracted elements of the communication stream message to an email message to be presented via an email system.

Abstract: Systems and methods are provided for reducing stream data according to a data streaming protocol under a multi-access edge computing. In particular, an IoT device, such as a video image sensing device, may capture stream data and generate inference data by applying a machine-learning model trained to infer data based on the captured stream data. The inference data represents the captured stream data in a reduced data size based on performing data analytics on the captured data. The IoT device formats the inference data according to the data streaming protocol. In contrast to video data compression, the data streaming protocol includes instructions for transmitting the reduced volume of inference data through a data analytics pipeline.

Abstract: A method performed by a node in a communications network for predicting energy usage in the communications network. The method comprises obtaining values of network parameters for a plurality of nodes in the communications network and corresponding energy usage measurements for the plurality of nodes, and obtaining a subset of the network parameters that are correlated with the energy usage measurements. The method then comprises training a first model using a first machine learning process to take as input the values of the subset of network parameters for the nodes in the plurality of nodes, and output a prediction of the corresponding energy usage measurements.

Abstract: Described are examples for using codelets executing within applications to use machine-learning (ML) models to infer a result based on application data. The codelets may be dynamically loaded into the applications during execution. A controller verifies, based on extended Berkeley packet filter (eBPF) bytecode of the codelet, that the codelet satisfies safety requirements for execution within the application. A computing device executing the application loads the verified codelet into a library of the application. The application executes the verified codelet to apply application data to the machine-learning model to infer a result. The ML model may be implemented by the eBPF code of the codelet or the codelet may include a call to a machine-learning model of a type supported by a controller of the application and a map for a serial representation of the machine-learning model. The computing device may reconstruct the ML model based on the serial representation.

Abstract: The present disclosure provides systems and methods that may advantageously apply machine learning to detect and ascribe network interruptions to specific components or nodes within the network. In an aspect, the present disclosure provides a computer-implemented method comprising: mapping a network comprising a plurality of islands that are capable of dynamically changing by splitting and/or merging of one or more islands, wherein the plurality of islands comprises a plurality of individual components; and detecting and localizing one or more local events at an individual component level as well as at an island level using a disaggregation model.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for identification of streaming activity and source for cached media on streaming devices.

Abstract: A system and method assists a user in achieving one or more goals of the user or the operator of a web site, application or other media by comparing a series of two or more actions of a user to those of other users, and identifying whether a primary action has been requested or performed by the user, or determining if a first piece of information has been specified by the user as being in a first set, and a second piece of information has not been specified or has been specified in a non-committal manner, and then provides relevant information to the user to instruct the user how to perform the primary action or to notify the user of the consequences of not specifying the second piece of information or specifying it in a non-committal manner.

Abstract: The invention relates to a quality of service (QoS)-aware parallel flexible Skyline service discovery method, including: performing, by each parallel node, a tournament selection algorithm to trim candidate services inside the parallel node; then making each parallel node perform a block dominance algorithm to filter candidate services inside the parallel node to obtain local Skyline services; collecting all local Skyline services to form a first set, and then performing filtering on the first set by using the block dominance algorithm to obtain global Skyline services; and determining whether there is a first constraint in the global Skyline services, and if yes, performing processing by using a flexible Skyline service correction algorithm to obtain a flexible Skyline service set for output, or otherwise, directly outputting the global Skyline services.

Abstract: The present disclosure provides systems and methods that may advantageously apply machine learning to detect and ascribe network interruptions to specific components or nodes within the network. In an aspect, the present disclosure provides a computer-implemented method comprising: mapping a network comprising a plurality of islands that are capable of dynamically changing by splitting and/or merging of one or more islands, wherein the plurality of islands comprises a plurality of individual components; and detecting and localizing one or more local events at an individual component level as well as at an island level using a disaggregation model.