Abstract:
A method includes: receiving M scrambled models from M online service nodes; receiving descrambling secrets from T online service nodes, where a descrambling secret of any online service node is determined based on M first shared keys of the online service node corresponding to the M online service nodes and N−M second shared keys of the online service node corresponding to N−M offline service nodes; and determining an aggregated model, where disturbance vectors of the M scrambled models are determined based on the descrambling secrets from the at least T online service nodes, at least N−T+1 first subsecrets determined by each service node based on N first shared keys and a first disturbance item corresponding to the service node, and at least N−T+1 second subsecrets determined by the service node based on N second shared keys and a second disturbance item corresponding to the service node.
Abstract:
An event processing method is provided, applied to filtering of events in a server. The method includes: matching a to-be-filtered event with an event matching rule, where the event matching rule is used for performing filtering to obtain an event that meets one or more first conditions and one or more second conditions, and the second condition indicates duration or repeatability of the one or more first conditions; and sending an instruction to a destination address when the to-be-filtered event meets the event matching rule. The event processing method provided in this application supports condition rule configuration of duration or repeatability, for example, a timing rule or a count rule.
Abstract:
This application provides a network function creation method and a communication apparatus. The network function creation method includes: A first entity receives a first request message, where the first request message is used to request to create or select a user network corresponding to a terminal device, and the first request message includes a user identifier of the terminal device. The first entity creates or selects a control plane function based on the user identifier, where the control plane function is used to create or select a user plane function corresponding to the control plane function, and the user plane function is used to support a subscription service corresponding to the user identifier. According to the network function creation method, the control plane function can be created, and the user plane function meeting all session subscriptions of a user can be created by using the control plane function.
Abstract:
This disclosure provides for methods and system for meta few-shot class incremental learning. According to an aspect a method is provided. The method includes obtaining at least one weight attention map of a first network and updating weights of a second network using the at least one weight attention map, where the second network is a modulatory network. The method further includes generating at least one feature attention map of the second network based on the at least one weight attention map of the first network and a set of input images of at least one class. The method further includes generating at least one feature map of the first network based on the set of input images of the at least one class, and updating the at least one feature map of the first network based on the feature attention map of the second network.
Abstract:
This application provides a data recovery method and apparatus, and relates to the field of computer technologies. The method includes: determining first data, where the first data is application data generated when an application program runs normally; storing the first data in a handle record, where the handle record is managed by using a reference count mechanism; and when the application program is unexpectedly restarted, obtaining the handle record, and restoring the application program, based on the handle record, to a status before the application program is unexpectedly restarted.
Abstract:
A system and a method for performing 2D on-screen user gaze estimation using an input facial image of a user captured using a camera associated to a processing device having a display. The method and system allow automated user calibration through automatic recording of calibration samples each including a calibration facial image of the user and an interaction point corresponding to a point on the display where an occurrence of a user interaction was detected when the corresponding calibration image was captured. The system and method also optimize user-specific parameters using the calibration samples by iteratively minimizing a total difference between the interaction points of a plurality of the calibration samples and corresponding 2D gaze estimation results and convert an estimated 3D gaze direction into a 2D gaze estimation result corresponding to a point on the display, by applying the users-specific parameters.
Abstract:
Embodiments of the disclosure provide a pool element (PE) status information synchronization method. A pool register (PR) receives a first registration message sent by a first PE, where the first registration message is used to add the first PE to a pool; the PR receives a second registration message sent by a second PE, where the second registration message is used to add the second PE to the pool; the PR determines a role of the first PE in the pool according to an identifier of the first PE and an identifier of the second PE, where the role is an active PE or a standby PE, and sends a status synchronization manner, the identifier of the first PE, and information about the role of the first PE to the second PE.
Abstract:
This application provides a network resource management method and a communication apparatus. The network resource management method includes: A first network element receives a first message from a second network element. The first message carries an identifier of a first terminal device, and the first message is used to request to create or select a first user service node corresponding to the first terminal device. Further, the first network element creates or selects the first user service node based on the identifier of the first terminal device, where the first user service node is configured to provide a network service for the first terminal device.
Abstract:
A method has the steps of obtaining a set of training samples from one or more domains, using the set of training samples to query a plurality of artificial-intelligence (AI) models, combining the outputs of the queried AI models, and adapting a target AI model via knowledge distillation using the combined outputs.
Abstract:
An optical storage medium, a method for preparing an optical storage medium, and a system are provided. In this application, the optical storage medium includes a substrate and at least one data layer, the data layer includes a recording layer and a spacer layer, the recording layer is located on the spacer layer, and the data layer is located above the substrate. The recording layer may store data, the recording layer includes an area in which a phase change material is distributed and an area in which no phase change material is distributed, and the two different areas may indicate different data.