Abstract: Learning-based service migration in mobile edge computing may be provided. First, a service migration policy may be created for a network that includes a plurality of edge clouds configured to provide a service to users. Next, a movement of a user receiving the service from a source edge cloud may be detected. The source edge cloud may be associated with a first area and the detected movement may be from the first area to a second area. Then, the service migration policy may be applied to determine whether to migrate the service for the user from the source edge cloud. In response to determining to migrate the service, a target edge cloud may be identified and the service for the user may be migrated from the source edge cloud to the target edge cloud. The service migration policy may then be updated based on a success of the migration.

Abstract: Learning-based service migration in mobile edge computing may be provided. First, a service migration policy may be created for a network that includes a plurality of edge clouds configured to provide a service to users. Next, a movement of a user receiving the service from a source edge cloud may be detected. The source edge cloud may be associated with a first area and the detected movement may be from the first area to a second area. Then, the service migration policy may be applied to determine whether to migrate the service for the user from the source edge cloud. In response to determining to migrate the service, a target edge cloud may be identified and the service for the user may be migrated from the source edge cloud to the target edge cloud. The service migration policy may then be updated based on a success of the migration.

Abstract: Learning-based service migration in mobile edge computing may be provided. First, a service migration policy may be created for a network that includes a plurality of edge clouds configured to provide a service to users. Next, a movement of a user receiving the service from a source edge cloud may be detected. The source edge cloud may be associated with a first area and the detected movement may be from the first area to a second area. Then, the service migration policy may be applied to determine whether to migrate the service for the user from the source edge cloud. In response to determining to migrate the service, a target edge cloud may be identified and the service for the user may be migrated from the source edge cloud to the target edge cloud. The service migration policy may then be updated based on a success of the migration.

Abstract: A method including building a relevancy model based on items in an item catalog. The method also can include building a title interpreter model based on titles of the items in the item catalog. The method additionally can include retrieving target attribute values that are associated with a target attribute name of a target item in the item catalog. The target attribute values can be received from multiple sources. The method further can include generating a respective relevancy score for each one of the target attribute values using the relevancy model. The method additionally can include generating a respective precision score for the each one of the target attribute values based on the title interpreter model. The method further can include determining a respective weight for the each one of the target attribute values based on the respective relevancy score for the each one of the target attribute values and the respective precision score for the each one of the target attribute values.

Abstract: Learning-based service migration in mobile edge computing may be provided. First, a service migration policy may be created for a network that includes a plurality of edge clouds configured to provide a service to users. Next, a movement of a user receiving the service from a source edge cloud may be detected. The source edge cloud may be associated with a first area and the detected movement may be from the first area to a second area. Then, the service migration policy may be applied to determine whether to migrate the service for the user from the source edge cloud. In response to determining to migrate the service, a target edge cloud may be identified and the service for the user may be migrated from the source edge cloud to the target edge cloud. The service migration policy may then be updated based on a success of the migration.

Abstract: A distributed controller of local radio resources implements a hybrid FFR system. The distributed controller provides user channel information to a central controller of regional radio resources. The distributed controller also obtains an FFR plan from the central controller. The FFR plan designates for an access point associated with the distributed controller, a central frequency band, an edge frequency band, and a distribution of user devices connected to the access point. The distribution indicates whether each user device communicates with the access point via the central frequency band or via the edge frequency band. The distributed controller provisions the access pint to connect to the user devices according to the FFR plan, and adjusts the distribution of user devices locally based on a change in the user channel information.

Abstract: A distributed controller of local radio resources implements a hybrid FFR system. The distributed controller provides user channel information to a central controller of regional radio resources. The distributed controller also obtains an FFR plan from the central controller. The FFR plan designates for an access point associated with the distributed controller, a central frequency band, an edge frequency band, and a distribution of user devices connected to the access point. The distribution indicates whether each user device communicates with the access point via the central frequency band or via the edge frequency band. The distributed controller provisions the access pint to connect to the user devices according to the FFR plan, and adjusts the distribution of user devices locally based on a change in the user channel information.

Abstract: Predictive scheduling may be provided. First, a first device may identify when a service flow is expected to become active. The first device may estimate an initial traffic profile in response to identifying when the service flow is expected to become active. The first device may then grant allocation based on the initial traffic profile of the service flow. Next, the first device may collect feedback to later update the traffic profile estimate. The first device may then update the traffic profile estimate.

Abstract: When displaying a frame that is to be transformed based on a predicted view orientation, the frame is transformed based on a predicted view orientation by, for each of plural different regions of the frame, determining a predicted view orientation transformation matrix to use to transform the region of the frame by interpolating between a first predicted view orientation matrix that corresponds to a transformation for a predicted view orientation for a first time during the display of the frame and a second predicted view orientation matrix that corresponds to a predicted view orientation for a second, later time during the display of the frame; and then using the interpolated predicted view orientation transformation matrix determined for the region of the frame to transform the region of the frame. In this way each different region of the frame is subjected to a different predicted view orientation transformation.

Abstract: When displaying a frame that is to be transformed based on a predicted view orientation, the frame is transformed based on a predicted view orientation by, for each of plural different regions of the frame, determining a predicted view orientation transformation matrix to use to transform the region of the frame by interpolating between a first predicted view orientation matrix that corresponds to a transformation for a predicted view orientation for a first time during the display of the frame and a second predicted view orientation matrix that corresponds to a predicted view orientation for a second, later time during the display of the frame; and then using the interpolated predicted view orientation transformation matrix determined for the region of the frame to transform the region of the frame. In this way each different region of the frame is subjected to a different predicted view orientation transformation.

Abstract: Systems, methods, and computer-readable media for coordinating scheduling between a wireless scheduler and a backhaul scheduler to reduce uplink data transmission latency. In some examples, an uplink data prediction for uplink data to transmit over a wireless connection of a wireless network is identified by a wireless scheduler of a wireless network. The uplink data prediction can be provided to a backhaul scheduler of a backhaul of the wireless network. The wireless scheduler can be coordinated with the backhaul scheduler by the backhaul scheduler to coordinate transmission of the uplink data through the backhaul based on the uplink data prediction. The uplink data can then be received at the backhaul for transmission through the backhaul based on the uplink data prediction.

Abstract: Predictive scheduling may be provided. First, a first device may identify when a service flow is expected to become active. The first device may estimate an initial traffic profile in response to identifying when the service flow is expected to become active. The first device may then grant allocation based on the initial traffic profile of the service flow. Next, the first device may collect feedback to later update the traffic profile estimate. The first device may then update the traffic profile estimate.