Abstract: Shared translation page tables are used for ownership transfer between software components having different execution privilege levels in a layered design. Each page table entry in a page table for the last translation level can include an ownership indicator to indicate the ownership of a physical page being mapped by the given page table entry by one or more software components. A single page table walk of the shared translation page tables can be performed to change the ownership of the physical page from one software component to another software component by updating the ownership indicator in a page table entry corresponding to that physical page.

Abstract: A hypervisor comprising a plurality of emulation routines is launched at a host. Individual emulation routines are used to perform respective tasks by a hypervisor on behalf of guest virtual machines of the host. A guest virtual machine initiates an operation which requires access to a virtual address. Subsequent to a determination by a memory management unit of the host that a virtual-to-physical address translation mapping for the virtual address which permits the operation initiated by the guest virtual machine is not available, a task is performed using a particular emulation routine, without initiating an analysis using the virtual address at the hypervisor to select the particular emulation routine from the plurality of emulation routines.

Abstract: Systems and methods are provided for generating a first model of a network based on a definition of the network, wherein the definition is used to deploy the network; generating a second model of the network based on a current deployment of the network; comparing the first model of the network and the second model of the network; and based on results of comparing the first model of the network and the second model of the network, generating output representing a discrepancy between the definition of the network and the current deployment of the network.

Abstract: This disclosure proposes systems and methods for processing natural language inputs using data associated with multiple language recognition contexts (LRC). A system using multiple LRCs can receive input data from a device, identify a first identifier associated with the device, and further identify second identifiers associated with the first identifier and representing candidate users of the device. The system can access language processing data used for natural language processing for the LRCs corresponding to each of the first and second identifiers, and process the input data using the language processing data at one or more stages of automatic speech recognition, natural language understanding, entity resolution, and/or command execution. User recognition can reduce the number of candidate users, and thus the amount of data used to process the input data. Dynamic arbitration can select from between competing hypotheses representing the first identifier and a second identifier, respectively.

Abstract: Systems and methods are disclosed for automated loading and packing for custom-sized packages. In one embodiment, an example system may include a first custom-sized package generation machine configured to generate packaging having a first range of dimensions, a second custom-sized package generation machine configured to generate packaging having a second range of dimensions, and a first independent transport robot configured to transport items to at least one of the first and second custom-sized box generation machines. The first custom-sized package generation machine and second first custom-sized package generation machine are decoupled from an item induction portion of the system.

Abstract: A content management system is provided. The content management system (CMS) uses a unique content modeling architecture using nested, reusable data entities, as well as a method for constructing a system around those entities to enable features and automation within the CMS. The solution's novelty lies in (1) a mechanism for efficiently and robustly governing, triggering, and enabling complex processes (e.g., translation workflows, publication workflows) for content within a CMS by dynamically associating system models with the content, and (2) enabling schema consistency and the creation of flexibly typed content via “model set” entities that aggregate one or more reusable content schemas.

Abstract: Systems and methods for performing malware scanning for a service provider network are disclosed. In response to accessing one or more files, file attributes may be determined. Confidence values may be generated based on the file attributes and may be used to select a scan operation. Such scan operations may include a scan operation using a 3rd party malware scan algorithm or a scan operation using a machine learning generated model. The selection of the scan operation may be performed based on a deferral learning model.

Abstract: Systems and methods are disclosed for forced air package redirection for improved fill rates. In one embodiment, an example sortation system may include a diverter configured to guide packages into a container, a first air delivery device configured to direct air at the package, a sensor configured to image the container, and a controller configured to determine, using the sensor, a position for a first package to be positioned in the container, determine the first package is adjacent to the first air delivery device, and cause the first air delivery device to be actuated, such that the air is directed at the first package. The air impacts the first package and causes the first package to change a direction or trajectory of movement of the first package.

Abstract: Devices and techniques are generally described for guided navigation into geofences. In various examples, a GUI of an application executing on a mobile device may display a first geofence surrounding a first delivery or pickup location. A reported location of the mobile device may be displayed on the GUI. A first navigation direction indicating a direction in which to walk to enter the first geofence may be determined. A first directional indicator may be displayed on the GUI indicating the first navigation direction. A determination may be made that the mobile device has entered the first geofence. First confirmation data indicating that a delivery to the first delivery address has been completed may be generated by the mobile device based at least in part on the mobile device having entered the first geofence.

Abstract: A media playback system may include features for mitigating interference. The media playback system may include a display device such as a smart TV that wirelessly transmits audio data to one or more listening devices such as earbuds, headphones, etc. The media playback system may include a remote control that can communicate using the same or similar frequencies and/or channels used for transmitting audio data. The media playback system may therefore use the radio(s) of the remote control to measure interference as a position removed from the smart TV and potentially closer to the listening device. This measurement may better reflect the listening device's ability to receive data packets reliably than a measurement of interference taken only at the location of the smart TV.

Abstract: Described herein is a computer-implemented method for generating a task-specific simulated robotic manipulator. A computing device receives, via a robotic design user interface, at least one of task information or constraint information. The computing device selects a set of components for a simulated robotic manipulator from a plurality of components in accordance with at least one of the task information or the constraint information. The computing device generates a simulated representation of the simulated robotic manipulator based at least in part on set of components and generated information. The computing device provides, for presentation via a display portion of the robotic design user interface, the simulated representation of the simulated robotic manipulator.

Abstract: Techniques for customer-initiated virtual machine resource allocation sharing are described. A hardware virtualization service of a cloud provider network receives a request to launch a first virtual machine, wherein the first virtual machine is of a first virtual machine type, the first virtual machine type having a resource amount allocated to virtual machines of the first virtual machine type. The hardware virtualization service causes a launch of the first virtual machine on a host computer system of the cloud provider network. The host computer system shares an allocation of the resource amount from a corresponding resource of the host computer system between the first virtual machine and a second virtual machine, wherein the second virtual machine is of the first virtual machine type.

Abstract: Systems and methods for robotic re-stowing of relocated inventory are described. In some examples, a determination may be made that a first tote located in a first zone of a processing facility includes a first item bound for a second zone of the processing facility. The first tote may be scanned by a first scanner of a first conveyance system to determine that the first tote is bound for the second zone. The first conveyance system may transport the first tote from the first zone to the second zone. The first tote may be robotically stowed on a first storage pod in the second zone. A first laden robotic drive unit may transport the first storage pod to a first pick station in the second zone.

Abstract: Systems and methods are disclosed for automated problem solve workflows for shuttle rail systems and related container movement devices. In one embodiment, an example system may include a shuttle and a controller. The controller may be configured to determine that the shuttle has a first item onboard, where the shuttle is configured to deliver the first item to a first location, determine that the first item was not delivered to the first location, and cause the shuttle to be routed to a problem solve station. The controller may be configured to determine a first container at which the shuttle is to unload the first item at the problem solve station, and cause the shuttle to unload the first item at the first container.

Abstract: A resource request is received by a peripheral device from host processing logic. The resource request includes a requested resource size. The peripheral device allocates resource of the peripheral device in response to the resource request. A resource response is sent by the peripheral device to the host processing logic. The resource response includes information indicating a location of the allocated resource.

Abstract: A system configured to improve encryption architecture for camera video data. By using an unbalanced binary tree architecture, the system enables group encryption that prioritizes a constrained device (e.g., embedded camera device), thereby reducing an amount of processing required for the constrained device to encrypt video data. For example, placing the constrained device near the root of the unbalanced binary tree offloads processing from the constrained device to other group members. At each device operating as a member of the group, a privacy enforcement policy operates to authorize/authenticate messages, using time-sensitive encryption keys that enable refined control over decrypted content. In some examples, the architecture can enable end-to-end encryption by removing any cloud services from the group. Alternatively, the architecture can include a cloud service with high or low privileges, enabling refined control over decrypted content while providing additional functionality.

Abstract: A method for modifying recorded audio of a video title may include: identifying a video title available for presentation; identifying a portion of audio data, of the video title, representing dialogue of an actor of the video title; identifying a first utterance of the dialogue to be replaced in the video title; selecting, based on a replaceable amount of the audio data, a second utterance to replace the utterance word in the portion of the audio data; generating, based on a voice profile of the actor, first speech signals representing the actor uttering the second utterance; removing second speech signals representing the actor uttering the first utterance from the dialogue; adding the first speech signals into the dialogue using at least a portion of the replaceable amount; and generating a modified version of the video title comprising the dialogue with the speech signals added.

Abstract: On-demand encryption may be implemented for online database encryption transition. Determinations can be made for individual data blocks as to whether individual data blocks have been encrypted using a new encryption scheme. Therefore read requests, write requests, or background operations can access and encrypt, if not already performed, different data blocks for the database while the database is in transition to the new encryption scheme.

Abstract: Systems and methods are provided to output audio data based on metadata associated with the audio data. A system can provide a buffer command to a plurality of computing devices. Based on providing the buffer command, the system can obtain, from the plurality of computing devices, a plurality of sets of audio data and a plurality of sets of metadata. The system can identify, for each time period of a plurality of time periods, a respective audio frame from the plurality of sets of audio data based on the plurality of sets of metadata. The system can generate a continuous audio stream based on identifying the respective audio frames and can route an output based on the continuous audio stream to a computing system.

Abstract: Techniques for a service provider network to identify malicious activity associated with a client device or service are discussed herein. A system can analyze network data and identify a device or service vulnerable to exploitation, disruption, data theft, and the like, by a threat actor operating as the IP address. In some examples, the system can determine a state for the network data based on a number of accounts and/or an amount or rate of new accounts accessed by the IP address over a time period. The system can compare states representing network activity by the IP address for different times and output an indication of the potential malicious event based on the compared states revealing a number of new resources accessed by the IP address meets or exceeds a threshold value.

Abstract: Speech processing tasks may be allocated at least partly to a local device (e.g., user computing device that receives spoken words) and at least partly to a remote device to determine one or more user commands or tasks to be performed by the local device. The remote device may be used to process speech that the local device could not process or understand, or for other reasons, such as for error checking. The local device may then execute or begin to execute locally determined tasks to reduce user-perceived latency. Meanwhile, the entire media input, or a portion thereof, may be sent to the remote device to process speech, verify the tasks and/or identify other user commands in the media input (or portion thereof).

Abstract: Techniques for text generation refinement are described. In some examples, a predicted text output is generated, wherein the predicted text output is selected from a plurality of predicted texts generated in response to the request, wherein the generating comprises: predicting a plurality of predicted texts using at least one machine learning model, scoring each of the plurality of predicted texts using at least one of intra-sample scoring and inter-sample scoring, computing a composite score for each of the plurality of predicted texts from the scores, ranking the plurality of predicted texts based at least in part the composite score, and selecting at least a highest ranked predicted text of the plurality of predicted texts as the predicted text output.

Abstract: Systems, methods, and devices are disclosed for front-lit displays having uniform brightness. In one embodiment, an example display may include an electrophoretic display, a light guide configured to direct light from one or more light emitting diodes, and a cover lens assembly. The cover lens assembly may include a cover glass layer, an anti-glare film coupled to the cover glass layer, and a hot melt adhesive disposed about lateral edge surfaces of the cover glass layer and the anti-glare film, such that the hot melt adhesive forms a perimeter of the cover lens assembly.

Abstract: Systems, apparatuses, and methods are described for delegated device authentication. An example method includes receiving, from a first device, a request to access enterprise data managed by an enterprise service. The example method also includes determining a security status of the first device, where the security status indicates that the first device complies with a set of conditional access policies. The example method also includes determining, based on execution of a first authentication mechanism having a first confidence rating, an account associated with a natural language processing system. The example method also includes receiving, from a second device, an authentication response indicating an execution of a second authentication mechanism having a second confidence rating and authenticating the first account based on the authentication response.

Abstract: Techniques for controlling robot movement based on identifier encoding signals are described herein. For example, a device can determine a safety state of a physical location indicating a permission to transmit signals. The safety state can be associated with an operation that can be performed at the physical location. The device can be associated with the physical location. The device can transmit, at a first time based at least in part on the trigger, a first signal that indicates a first identifier. The first identifier can match a second identifier available to an autonomous mobile robot (AMR) from a location marker associated with the physical location. The device can transmit, at a second time, a second signal that indicates the first identifier, the first signal and the second signal enabling the AMR to perform the operation.

Abstract: Universal Verification Methodology (UVM) for testing an integrated circuit includes receiving a configuration object having an address map code. A template code is enabled to be associated with the testing and it includes at least specifications associated with the UVM. A utility module is compiled to provide a model and a testbench environment within the UVM, where the model and the testbench environment are represented by a first class script and that is based in part on input from the template code and from the configuration object. A change is determined in the configuration object and a second class script is automatically generated using the utility module and based in part on the change in the configuration object, where the specifications remain unchanged for the second class script. The integrated circuit is tested using the second class script.

Abstract: Devices and techniques are generally described for parallelizing determination of image grayscale pixel values based on their color filter array locations in e.g., a frame buffer associated with an electrophoretic display device. An example method may include receiving an image stored in a frame buffer associated with an electrophoretic display of a computing device, wherein the electrophoretic display includes black and white ink and a color filter array overlaid thereon; identifying, for the first image, a first processing region of the frame buffer, a second processing region of the frame buffer, and a third processing region of the frame buffer; determining, using a serial processing operation, grayscale values for pixels in the first processing region; determining, using a parallel processing operation, grayscale values for pixels in the second processing region; and determining, using the serial processing operation, grayscale values for pixels in the third processing region.

Abstract: Systems and methods for acoustic event detection customization include determining one or more acoustic event detection (AED) models to send to a device based on enrollment data, AED model usage data, and/or other data indicating a likelihood that an acoustic event will be detected utilizing the one or more AED models. Additionally, an activation schedule may be generated and utilized to determine when to activate and deactivate the one or more AED models.

Abstract: Techniques for traffic shaping are described. In some examples, an ingress point receives traffic from a source and determines to send the traffic to a traffic shaper instance. The shaper instance is capable of performing one or more of a plurality of actions, wherein the plurality of actions include altering a shape of the traffic prior to sending the traffic to an original destination of the traffic, re-directing the traffic to a honeypot, allowing the traffic without doing altering action, and blocking the traffic.

Abstract: A labeled data set which indicates layout elements within document images, as well as a reading order for the layout elements, is obtained. Using the data set and a combined loss function, a machine learning model is trained to predict respective content classes of layout elements identified within document images, and reading orders for the layout elements. The combined loss function aggregates losses associated with prediction of the content classes and the reading order. In response to an inference request indicating a document image, an indication of a predicted reading order for layout elements identified in the document image is generated using the trained model.

Abstract: Systems and methods are provided for an on-demand code execution service configured to receive, from a virtual computing component executing on a host computing device, a communication that a first process of an application of the virtual computing component has completed execution, wherein the virtual computing component is allotted a set of computing resources for execution; send, to the virtual computing component, a notification indicating that a computing resource allotted to the virtual computing component will be de-allocated; receive, from the virtual computing component, a request to delay de-allocation of the computing resource, wherein a second process of the application executes subsequent the virtual computing component sending the communication; and delay de-allocation of the computing resource based on the request from the virtual computing component.

Abstract: In a multi-tenant network, event routing and encryption techniques involve processing events through an event bus service. When an event is received, it is evaluated against routing rules. If the event matches a rule linking to a resource in another customer account, the event data is encrypted using a key associated with that account. Finally, the encrypted event is delivered to the target resource, ensuring secure communication between different customer accounts within the network.

Abstract: Systems and methods are described for monitoring bandwidth use of code executed on a serverless execution environment on a per-execution basis. The code—which may correspond to a serverless function—is illustratively executed in an execution environment provided by a host computing device, which environment is configured to support a single execution of the function's code at a given time. Prior to an execution, the host computing device can initiate tracking of bandwidth use of the execution environment. The host computing device can then execute the code within the environment and, subsequent to such execution, finish tracking of the bandwidth use. The total bandwidth use observed based on tracking can then be reported as bandwidth use of the code execution.

Abstract: A query may be selected for asynchronous processing, between synchronous query processing and asynchronous query processing. Asynchronous processing may be performed by a query engine, a storage engine, and a circular buffer of the database. The query engine may send requests for records to the storage engine and process records from a readable portion of the circular buffer. The storage engine may obtain records from the database based on the received requests, and store records within a portion of the circular buffer locked to prevent the query engine from reading the portion of the circular buffer. A query result may be generated from the records ready from the circular buffer by the query engine.

Abstract: Systems are generally described for enabling electro quasistatic (EQS) communication for devices. An example system includes a first electronic device comprising one or more first processors, a first converter circuitry coupled to at least one first processor, a first electrode, coupled to the first converter circuitry, positioned in proximity to a metal structure of a third device. The example system also includes, a second electronic device comprising a second electrode, coupled to the second converter circuitry, positioned in proximity to the metal structure of the third device and configured to receive EQS communication from the first device. The example system also includes a third device comprising the metal structure. The first device may be configured to encode a first datagram as a modulated signal waveform, and transmit by the first electrode, the modulated signal waveform via electro quasistatic transmission through the metal structure.

Abstract: Technology for selecting a control channel is described. Historical radar event data associated with one or more dynamic frequency selection (DFS) channels is used to determine a first score and a second score for a first set and a second set of channels, respectively. The first and second scores are used to determine that a radar event is less likely to occur on the first set than the second set of channels. A bonded channel is selected to include the combination of channels on which a radar event is less likely to occur. Historical interference data is used to determine a third score for a first channel of a bonded channel that includes at least the first set of channels. The third score is used to determine a control channel from the channels of the bonded channel. Management information associated with the bonded channel is sent on the control channel.

Abstract: Dimension reduction is performed for multimodal embedding searches. A request to perform a search using input data of a modality of a multimodal machine learning model is received. The input data is encoded using the multimodal machine learning model to generate a first embedding that represents the input data. A machine learning dimension reduction technique is applied to the first embedding to generate a second embedding with a reduced number of dimensions. This second embedding is then used to search an index of data objects also generated using both the multimodal machine learning model and the machine learning dimension reduction technique.

Abstract: In response to registration requests, metadata of a plurality of machine learning artifacts are stored. In response to a pipeline definition request, which does not specify resources to be used for the pipeline, a representation of a pipeline comprising nodes corresponding to registered artifacts is stored. A benchmarking operation is conducted to select the types of resources to be used for the nodes, and results of the benchmarking are provided.

Abstract: A traffic controller (TC) circuit monitors an ingress stream being fed to an integrated circuit (IC) component comprising the TC circuit, wherein the IC component is part of a processing pipeline of a computing device. The TC circuit controls input traffic on the ingress stream to processing logic circuitry of the IC component. The TC circuit also controls output traffic on an egress stream being outputted from the IC component. The TC circuit generates an interrupt based on a set of operational metric configurations, and the ingress stream or the egress stream.

Abstract: Techniques for disconnected remote task restarts via situational distributed control are described. A container instance operates in a cluster of container instances that is remote from a control plane. Each agents of the container instances can be configured with a control module that provides purposefully limited control plane functionalities that are activated in specific scenarios. Each agent can be configured, by the control plane, with mappings indicative of other agents—and their currently-executing tasks—that the agent is to serve as a backup control plane for. Upon a failure associated with an agent coupled with a lack of network connectivity from the cluster to the control plane, another agent can become a backup control plane to restart task execution for the affected agent.

Abstract: Techniques for providing graphic overlays on live videos using a depth machine learning (ML) model are described. According to some examples, a computer-implemented method includes receiving a live stream of a sporting event; determining a mapping of a playing surface of the sporting event depicted in the live stream to a frame of the live stream; generating, by a machine learning model, a depth map indicating a corresponding depth for an individual player of a plurality of players depicted in the frame based at least in part on an input of the mapping and an indication of the plurality of players depicted in the frame; and inserting a graphic into the frame of the live stream based at least in part on the depth map.

Abstract: In some disclosed embodiments, audio data corresponding to an utterance may be received and processed to determine at least a first natural language understanding (NLU) hypothesis and a second NLU hypothesis for the utterance. A determination may be made that the first NLU hypothesis is likely more likely accurate than the second NLU hypothesis, and a device may be caused to present content corresponding to the first NLU hypothesis. The device may further be caused to display a graphical user interface (GUI) element that is selectable to cause execution of an action corresponding to the second NLU hypothesis.

Abstract: Input data comprising video and associated audio is processed to generate recap data, such as video with audio that summarizes previous occurrence and entices a viewer. The input data is processed to generate dense text data that includes timestamps, dialogue, descriptions of scenes of the video, and so forth. The input data is also processed to generate summary data, such as scene, episode, and season summaries. A set of one or more machine learning (ML) systems use this as input to generate a narration script. Narration audio is generated from the narration script. The narration script is then used to generate sections of a recap such as an intro, teaser, detail, and outro. Each portion may include narration audio, clips of video, video clips of relevant dialog, background music, audio effects, and so forth. The duration of elements may be constrained based on the presentation of associated narration audio.

Abstract: Disclosed are various embodiments for dynamic deployment optimization for network functions in radio-based networks. In one embodiment, a deployment specification for a network function in a radio-based network is received. A particular node is dynamically selected from a plurality of nodes in a cloud provider network for executing the network function. The network function is deployed for execution on the particular node according to the deployment specification.

Abstract: Log records are collected from resources used for various constituent services of a distributed application during an execution of the application. An aggregated event record sequence corresponding to the execution is generated from the log records, with individual event records indicating occurrences of events for which expected responses of the constituent services are indicated in correctness checking rule sets associated with the application. An action responsive to a violation of a correctness checking rule set is initiated after the violation is detected based on analysis of the event records.

Abstract: Disclosed are various embodiments for an interactive user interface for predictive simulation results that are derived from the automatic sampling of a simulation space. In some embodiments, a system is configured to receive a manifest and an input data object for executing a simulation project on a simulator. The system can generate sample inputs for a sample space of the simulation project based on a sampling space criteria for the simulation project. A simulation of the simulation project is executed to generate simulation outputs based on the input data object. A machine learning model can be trained based at least in part on simulation outputs. The system can be configured to generate a user interface that displays a predictive simulator output for the simulation project based on querying the machine learning model with an unsimulated input.

Abstract: Devices and techniques are generally described for compression of machine learning models. In some examples, a first weight value wi of a machine learning model is determined. The first weight value wi may be associated with a first region comprising an interval of numbers. A first loss for the first weight value wi may be determined using a first loss function comprising a weighting term comprising a sinusoid function with an argument of ??rwi. ? may be selected such that a maximum value of the weighting term for the first region is of a format that is compatible with first machine learning accelerator. A gradient of the first loss function may be determined. A second weight value wi may be determined using the gradient and the first loss. The second weight value wi may be stored in a first memory.