Abstract: Techniques for self-tuning an analytics system via observed execution optimization are described. Upon a need for execution resources, a resource manager can select a type of executor from multiple candidate executor types based at least in part on one or more of current execution data associated with the execution of tasks of a user application and/or historic execution data associated with one or more other applications. The current execution data may include event log data originated by the driver application based on the execution of the user application and/or metric data describing characteristics of one or more worker nodes involved with executing the user application or characteristics of one or more other executors implemented by the one or more worker nodes in executing the user application.

Abstract: Code changes may be hierarchically clustered to discover coding practices. Code change graphs for changes to code in a source code repository may be clustered according to hierarchy of different features determined for the source code into groups. The code change graphs in the groups may then be indexed according their similarity with other code change graphs in the groups. Then one or more coding practices corresponding to the indexed code changes may be provided.

Abstract: Technologies directed to interactive physical placement of devices for optimal motion sensing using channel state information (CSI) are described. A method includes determining that a size of received CSI data exceeds a first threshold or is received in a specified time interval, the CSI data representing channel properties in a wireless network of devices. The method further includes validating a physical placement of the wireless devices by i) determining that a parameter value associated with the CSI data is less than a second threshold representing a no-motion condition and the first CSI data matches ground truth data associated with the no-motion condition; or ii) determining that the parameter value is greater than a third threshold representing a motion condition and the first CSI data matches ground truth data associated with the motion condition. The method sends to a user device an indication that the physical placement is validated.

Abstract: Layer-specific modifications may be applied to a network communication received at a proxy according to configurable rules at the proxy. A modification configuration may be received at the proxy that corresponds to a layer of a communication stack supported by the proxy. When a communication is received, data from the layer of the communication stack of the network communication may be evaluated to determine whether a modification is applicable to the communication according to the modification configuration. An applicable modification may be applied by proxy as part of processing the communication at the proxy.

Abstract: Systems and methods are provided for conducting continuous changelog-based replication of modifications made to a file system to a target storage system. An export system can monitor a changelog on a file system to detect changes made to the file system. The exporter system can further filter and reformat the changes for replication to the target storage system by storing the changes in a pending changes data store. To reduce duplicative traffic, the export system can store changes for a configurable delay period, writing all pending changes to a file system object after the delay while aggregating and deduplicating the changes, thus improving performance relative to immediate replication.

Abstract: Techniques for machine learning (ML) and natural language processing (NLP) are described. One technique enables the creation of a clean training dataset through just a few API calls. Another technique provides an automated process for generating a domain-specific lexicon, which is then used to generate ML training datasets, in a manner that requires little to no human labor. Another technique gathers ML training data from domain-specific public sources, which are more likely than typical public sources to contain focused terminology and to be free from errors, thus resulting in trained ML models that provide more accurate inferences.

Abstract: Systems and methods are described herein relating to ingesting data objects from various sources and associating them in a hierarchy with a profile. In some aspects, a data object, including a set of fields and indicating a data object type, may be obtained for associating with a profile maintained by a computing resource service provider. The data object may be converted into a standardized format based on the data object type indicated in the data object using an object mapping associated with the data object. A key associated with the data object and indicated in the object mapping may be selected and used to search an index for a related profile. Upon finding a profile with the matching key, the profile may be updated to include an association of the standardized data object, such as in a hierarchy.

Abstract: Technologies directed to a noise-immune miniaturized antenna (NIMA) structure in a main logic board (MLB) and diverting surface currents from the MLB to a metal structure to reduce noise coupling from a chipset on the MLB to the NIMA structure are described. The NIMA structure is located at a side of the MLB and includes a first tuning component coupled to a distal end of a radiating arm of the NIMA structure and a second tuning component coupled to a distal end of a shorting arm of the NIMA structure. The NIMA structure radiates in a first frequency range and a second frequency range. A conductive fastener couples the MLB to a metal structure to divert surface currents from the MLB to the metal structure.

Abstract: Described herein is a system for tracking objects and performing dynamic entity resolution using image data. For example, the system may build an environment map and populate the map with objects present in the environment. As the devices move about the environment it may capture image data and, based on its position and/or configuration of its components, may determine updated locations of objects that move in the environment. Upon receiving a query from a user, based on the location of the objects relative to the device/user, the system can interpret gestures and voice commands to infer which object is specified by the voice command. To build the environment map, the system performs object detection to generate bounding boxes associated with an object, then clusters the bounding boxes into a three-dimensional (3D) object associated with 3D coordinates. As the system tracks the object using the 3D coordinates while maintaining two-dimensional (2D) information (e.g.

Abstract: Unstructured data items are stored at an object storage service. A filtering requirement to be used to generate a result set for an access request is determined. Using a transformed representation of the filtering requirement, a target set of tokens of the filtering requirement which are to be obfuscated within a log record is identified. A log record that comprises substitute tokens for the target set of tokens is generated and stored.

Abstract: A provider network implements service discovery and renaming. A provider network includes a domain name service (DNS) and a renaming tool. The DNS receives queries for service names and sends query logs to a data store. Each query log indicates the queried name and a network address (e.g., IP address). A user may request that the service be renamed to a new name. In response, the renaming tool adds a new entry to for the new name to the DNS. The renaming tool retrieves data from the query logs and identifies clients of the service. The renaming tool sends a notification to the identified clients that the name has changed to the new name. The renaming manger retrieves additional data from the query logs to identify how many of the clients have switched to using the new name. After enough clients have switched, the old name may be decommissioned.

Abstract: An IoT gateway at a client site uses sensing functions and actuating functions to safely implement actions for equipment. The client provides, to the IoT service, specifications for a sensing function and one or more actuating functions that are to communicate with the sensing function. The functions are deployed to the IoT gateway at the client site. The sensing function may be assigned read only access to the equipment and the actuating function may be assigned write only access to the equipment. The sensing function may collect data from the equipment and determine that a condition has been met (e.g., temperature is above an upper limit). In response, the sensing function provides an indication of an action to the actuating function (e.g., via a unique topic). If the action is a supported action, then the actuating function causes the equipment to perform the action; otherwise, it ignores the indicated action.

Abstract: An airframe design may include a frame or assembly, and one or more components that may be removably attached to the frame. The frame may include struts, central bulkheads, a tail section, a plurality of wing sections, and motor mounts that are attached together. The one or more attachable components may include a forward fuselage, motors, propellers, motor pod fairings, stabilizer fins, and landing gear that are attached using fasteners. The frame may reduce the number of parts of the airframe design and may also reduce complexity, cost, and weight, while also increasing stiffness and strength. Further, the various attachable components may facilitate fabrication, assembly, and maintenance of an aerial vehicle having the airframe design.

Abstract: Systems and methods for training and performing query answering for programming related query answering are disclosed. During a training phase, queries and answers of a training dataset may be encoded as vectors into the Euclidean space. Loss values for query-answer pairs, duplicate queries, and augmented queries may be determined using loss functions and may be used to adjust the model of the encoder. During the operation phase, a query may be received and encoded into the Euclidean space. The encoded answer most proximate to the encoded query may be decoded and provided as the most relevant answer.

Abstract: An item retrieval system can include a support defining a continuous pathway having an upper level, a lower level, and an end portion connecting the upper and lower portions. An advancement mechanism can move a carrier having a container advanced along the continuous pathway between the upper and lower levels.

Abstract: Queries may be accelerated with disjunctive equijoin conditions. A query is received and a disjunctive equijoin condition identified. Different strategies, such as creating a filter data structure for distinct equijoins may be implemented where the filter data structure is applied to the result of scan operations for the disjunctive equijoin conditions. These strategies may be incorporated into a query plan generated for the query, which is then executed to return a result for the query.

Abstract: A system configured to perform acoustic feedback control to enable a device to perform hearing enhancement while suppressing acoustic feedback. During hearing enhancement, the device may amplify environmental noise based on a unique hearing profile associated with the user, personalizing equalization settings, a dynamic range, and/or other characteristics to optimize playback audio for the user. The acoustic feedback control may include an acoustic feedback cancellation (AFC) component that uses an adaptive filter to estimate and cancel a feedback signal. In addition, the AFC component may perform entrainment prevention by detecting periodic signals and adjusting an adaptation rate of the adaptive filter accordingly. Separately, the device may selectively suppress acoustic feedback by detecting frequency bands representing acoustic feedback (e.g., squeal detection) and applying one or more notch filter(s) to suppress the selected frequency bands (e.g., squeal suppression).

Abstract: This disclosure describes a system for automatically updating item image information stored in an item images data store and used for processing captured images to identify items represented in those images. In one implementation, once an identity of an item has been verified, captured images of that item are associated with the item and stored in the item images data store. As a result, the item images data store is updated each time an image of the item is captured and the identity of the item is verified.

Abstract: Features related to systems and methods for reinforcement learning are described. The environment includes one or more agents for automating the training of reinforcement learning (RL) models. The environment may include a simulator or real-world observations. The features described identify key training parameters, resource configurations, virtual network configurations, and simulators based on historical training data.

Abstract: This disclosure describes a test prediction service for predicting needed resources and associated costs for testing projects related to testing of new programs, services, etc., within a service provider network. The test prediction service uses a first machine learning model for predicting “hard costs” using data from a first data source that includes end-to-end test details such as associated resource and infrastructure use (costs) of the service provider network during previous end-to-end testing projects. A second machine learning model is used for predicting “soft costs” using data from a second data source that includes data, e.g., engineering headcount/hours spent developing the end-to-end testing and/or during previous end-to-end testing. The data is used to train the machine learning models. The test prediction service uses the trained machine learning models to generate estimated testing costs of new testing projects when a user enters project attributes for a new testing project.

Abstract: Systems and methods are described for use of reserved storage capacity at a cloud provider network extension, sometimes referred to as a provider substrate extension (“PSE”) or “outpost”, as a data cache, thereby improving performance of the PSE. The reserved storage capacity may be reserved for a primary purpose, such as data replication in the event of failure of an in-use portion of the PSE's storage capacity. During non-failover operation, the reserved storage capacity may be otherwise unused. Embodiments of the present disclosure enable the reserved storage capacity, when not used for its primary purpose, to be used as a data cache to store information potentially used at the PSE.

Abstract: Techniques are disclosed for generating a representation of a motion of an object through a physical topography. In one example, a system receives from a camera a sequence of images that shows the motion of the object within a first field of view of the camera that shows a portion of the physical topography, the portion being mapped to a three-dimensional coordinate space. The object is detected using a computer vision model, and a motion profile of the object is determined that maps the motion of the object to the three-dimensional space. The motion profile is provided to a computing service for being combined with audiovisual data (e.g., including a three-dimensional model of the physical topography), and whereby the combined data is then provided to a user device for subsequent presentation.

Abstract: Devices and techniques are generally described for selection of an object to follow during a video call. In various examples, a first frame of image data may be received from a camera. First image data representing a first person and second image data representing a second person may be determined. A first priority level may be determined for the first person at a first time. A second priority level for the second person may be determined at the first time. The camera may be controlled to follow the first person based at least in part on the first priority level and the second priority level.

Abstract: Systems, devices, and methods are provided for authorizing access to operating system resources using security policies managed by a service external to the operating system. An operating system may be provisioned with a kernel-mode component that intercepts system calls from applications, determines a request context for the system call, and sends a request to an external policy management service. The policy management service may be used to perform a policy evaluation to determine whether to grant access to operating system resources. In some cases, policies are cached by the operating system. In various examples, the operating system and policy management service are both hosted on resources managed by a computing resource service provider on behalf of a customer to run mainframe workloads.

Abstract: Systems, devices, and methods are provided for an automated sleep assistant. An automated sleep assistant may comprise an audio input device for capturing sounds and utterances made by a user while sleeping. Audio data may be collected and analyzed to identify sleep patterns indicative of an adverse sleep-related condition such as Obstructive Sleep Apnea (OSA). A pillow comprising a plurality of regions with one or more sensors, one or more inflatable air bladders, and one or more haptic feedback devices may be utilized. A base unit device physically connected to the pillow may determine one or more external forces that mitigate the adverse sleep-related condition and cause the pillow to apply the one or more external forces to the user, for example, by using a pump to inflate an air bladder to help the user's head lift or tip to a specific angle that opens his or her airway.

Abstract: A local device may receive, from a remote device, a task to be completed at a workstation within a logistics center in which orders are fulfilled and processed. The local device may determine, among a plurality of robotic agents at the workstation, a robotic agent configured to perform the task. An in-memory datastore having cache and compute capabilities may be stored in memory of the local device may store data associated with the task and the robotic agent to complete the task. The local device may instruct the robotic agent to perform the task and may receive, from the robotic agent, an indication associated with the performance of the task. This data may be stored in the in-memory datastore and/or may be used for assigning future tasks.

Abstract: An enhancement equalizer (EEQ) can be configured to compensate for hearing loss. An app or application can assist a user with measuring hearing loss at various frequencies (e.g., threshold sensitivity versus normal hearing). Using these measurements, the system may compute a set of filters for an EEQ that can boost different frequencies by different amounts corresponding to the user's sensitivity to that frequency. The measurement and resulting EEQ may be earphone specific (e.g., both the measurement and the filter computation may depend on the particular type/model of earphone used). In some implementations, the system may allow the user to select correction strength that controls an amount of correction applied (e.g., 25%, 50%, or 75% of full correction). In some implementations, the system may adjust the EEQ and/or correction strength according to the volume of playback (e.g., by applying less correction at higher playback volumes to avoid triggering earphone limiters).

Abstract: One challenge for middle-mile route planning is that the set of loads changes significantly between daily planning and execution. Systems and methods are provided for optimizing a transportation plan for a transportation network based on these load changes. The disclosed system re-optimizes a solution by starting from a previously existing plan and previously generated columns (e.g., candidate routes). The disclosed techniques significantly improve the compute time of the system to generate transportation plans that are optimized accordingly to an optimization parameter. The system takes into account the current execution status associated with a given entry of the plan to determine whether the entry should be re-optimized. Entries corresponding to tours that have already commenced, may be at least partially ignored for re-optimization consideration. The disclosed techniques enable state-aware, adaptive re-optimization for even tours that are in-progress or have been tendered.

Abstract: Devices and techniques are generally described for cohort determination in natural language processing. In various examples, a first natural language input to a natural language processing system may be determined. The first natural language input may be associated with a first account identifier. A first machine learning model may determine first data representing one or more words of the first natural language input. A second machine learning model may determine second data representing one or more acoustic characteristics of the first natural language input. Third data may be determined, the third data including a predicted performance for processing the first natural language input by the natural language processing system. The third data may be determined based on the first data representation and the second data representation.

Abstract: In certain embodiments, a web services system receives a request to provision a device, such as a telephone, as an authentication device. The web services system initiates display of an image communicating a key to allow the telephone to capture the image and to send key information associated with the key. The web services system receives the key and determines that the key information is valid. In response to the determination, the web services system sends a seed to the telephone to provision the telephone to be an authentication device. The telephone can use the seed to generate one-time passcodes to access a service of the web services system.

Abstract: Systems and methods are disclosed for foldable containers with improved dumping features. In one embodiment, an example foldable container may include a first container wall, a second container wall coupled to the first container wall at a first joint, the second container wall configured to rotate outwards in a first direction with respect to the first container wall, and a third container wall coupled to the first container wall at a second joint, the third container wall oriented perpendicular to the first container wall. Example foldable containers may include a number of wheels and smooth side panels, where a first cross-sectional area of the foldable container at a lower portion of the foldable container is greater than a second cross-sectional area of the foldable container at an upper portion of the foldable container.

Abstract: Devices, systems, and methods are provided for brand matching using multi-layer machine learning. A method may include generating, based on a first embedding vector and a second embedding vector as inputs to a twin neural network, a third embedding vector and a fourth embedding vector; generating, based on the first embedding vector and the second embedding vector as inputs to a difference neural network, a difference vector indicative of a difference between the first embedding vector and the second embedding vector; generating a concatenated vector by concatenating the third embedding vector with the fourth embedding vector and the difference vector; generating, based on the concatenated vector as an input to a feedforward neural network (FFN), a score between zero and one, the score indicative of a relationship between a first entity and a second entity.

Abstract: Techniques for an efficient inter-prediction structure and signaling for low-delay streaming of live video are described.

Abstract: Systems and methods of determining and providing optimized deployment plans for deploying software to vehicles are disclosed. In some embodiments, a vehicle software deployment system evaluates one or more cost functions to determine relative costs of different deployment configuration options for deploying software to a vehicle, such as resource costs (e.g., bandwidth, compute, memory, etc.), isolation costs (e.g., limited access to input information, limited connectivity to other ECUs, etc.), performance costs, etc. Based on the evaluation of the one or more cost functions, the vehicle software deployment system determines an optimized deployment plan. Also, the vehicle software deployment system receives telemetry data from the vehicle and automatically determines updated optimized deployment plans in response to changes in configuration of the vehicle indicated in the telemetry data.

Abstract: A vehicle software test environment management system provides a virtual vehicle environment that includes virtual electronic control units (vECUs) having a virtual bus connectivity configuration used to simulate respective ones of electronic control units (ECUs) of a real-world vehicle. The vehicle software test environment management system determines respective instance types of one or more virtual compute instances to be used to implement the vECUs based on respective configuration of respective ones of the ECUs and further determines respective machine images to emulate respective software environments of the respective ones of the ECUs. The vehicle software test environment management system may also deploy a vehicle software application to be certified on one or more of the vECUs and test the deployed vehicle software application using recorded signals of one or more ECUs of the real-world vehicle.

Abstract: Based on analysis of a workload associated with a throttling key of a client request directed to a first service, a scale-out requirement of the throttling key is obtained at respective resource managers of a plurality of other services which are utilized by the first service to respond to client requests. The resource managers initiate, asynchronously with respect to one another, resource provisioning tasks at each of the other services to fulfill the scale-out requirement. A throttling limit associated with the throttling key is updated to a second throttling key after the resource provisioning tasks are completed by the resource managers, and the updated limit is used to determine whether to accept another client request associated with the throttling key.

Abstract: An automatic technique is disclosed to enrich presented answers by highlighting relevant shopping recommendations. The shopping recommendations can either be highlighted within the answer itself, or as an auxiliary list of suggestions. A model is described for selecting phrases from the answer text (sequences of consecutive terms called noun phrases) that refer to potential products that likely represent relevant shopping recommendation in context of the question-answer pair. The noun phrases are then ranked in order of importance. The top-ranked noun phrases are used to search products to be displayed in association with the noun phrases. Clicking or tapping on a highlighted noun phrase launches a shopping-related flow, such as presenting a widget with product recommendations or running a search in a search engine.

Abstract: Synchronous replication for a distributed database system may be performed using an erasure coding scheme. A request that causes a write to a database hosted in a distributed database system is received. A replication message for a synchronous replication technique is generated, then divided and encoded into a number of chunks according to an erasure encoding scheme that allows the replication message to be reassembled with less than the number of chunks. The chunks are sent to another instance of the database which receives and reassembles the replication message from the chunks and responds to acknowledge that the write is committed.

Abstract: A vehicle signal relay system enables a relay agent in a first zone of a vehicle to send sensor signals having a first link-layer communication protocol to a software application deployed on a compute unit in another zone of the vehicle that is connected using another link-layer communication protocol. The vehicle signal relay system allows the software application to identify target relay agents with access to needed sensor signals. The vehicle signal relay system may further enable one way or mutual attestation. The vehicle signal relay system may also allow filters to be applied to the subscribed vehicle sensor signals, and may allow the software application to determine a communication protocol to be used between the software application and the relay agent.

Abstract: Techniques for encoding quantum circuit mapping problems as SAT solver optimization problems are disclosed. Quantum circuit mapping often requires the use of SWAP gates in order to configure logical quantum computations to be executed using fixed quantum hardware device layouts. A quantum compilation service takes a logical quantum circuit, a physical qubit connectivity graph, and a requested number of SWAP gates to solve the mapping using and encodes the information into a Conjunctive Normal Form (CNF) equation using a layout-transition-based encoding scheme. The CNF equation is then provided to a SAT solver which attempts to determine an assignment for the mapping using the set number of SWAP gates requested. Multiple CNF equations corresponding to different requested numbers of SWAP gates may be solved for in parallel using multiple SAT solving instances.

Abstract: Systems and methods for performing medical audio summarizing for medical conversations are disclosed. An audio file and meta data for a medical conversation are provided to a medical audio summarization system. A transcription machine learning model is used by the medical audio summarization system to generate a transcript and a natural language processing service of the medical audio summarization system is used to generate a summary of the transcript. The natural language processing service may include at least four machine learning models that identify medical entities in the transcript, identify speaker roles in the transcript, determine sections of the transcript corresponding to the summary, and extract or abstract phrases for the summary. The identified medical entities and speaker roles, determined sections, and extracted or abstracted phrases may then be used to generate the summary.

Abstract: In response to a programmatic request, configuration information representing a multi-network-segment gateway established on behalf of a customer is stored at a networking service. In response to another programmatic request, a communication session is established between a route signaling node of the gateway and a routing information source located at a customer premise. In response to additional programmatic input, the networking service stores an indication that the gateway is to be used to transfer packets between a cloud-side virtual network and a customer-side virtual network. The routing information exchanged in the session pertains to the cloud-side and customer-side virtual network, and is used to transfer data packets between the two virtual networks.

Abstract: Techniques are described for executing satisfiability modulo theories (SMT) solvers in a “shadow” system configuration where input queries are provided to a primary SMT solver system and additionally to one or more secondary SMT solver systems. SMT solver systems can be used by cloud providers and in other computing environments to analyze the implications of configured user account policies defining permissions with respect to users' computing resources and associated actions within a computing environment, to help ensure the security of computing resources and user data, etc. The results generated by a primary SMT solver system can be provided to one or more secondary SMT solver systems, where each of the secondary SMT systems can comprise different system components or different versions of system components, to assess the correctness of the primary SMT solver system, to compare performance metrics, among other possible types of analyses.

Abstract: Systems and methods for providing vehicle software deployment plans that include one or more fallback deployment plans are disclosed. In some embodiments, a vehicle software deployment system determines a deployment plan for deploying one or more software applications one or more electronic control units (ECUs) of a vehicle. Additionally, for one or more failure scenarios, the vehicle software deployment system determines one or more respective fallback deployment plans, such as a deployment plan that assumes one of the ECUs of the vehicle has failed. A deployment plan bundle is provided to the vehicle, comprising a primary deployment plan as well as the one or more fallback deployment plans. In the event that one of the failure scenarios takes place with respect to the vehicle, a deployment agent of the vehicle automatically deploys one of the fallback deployment plans that was provided with the deployment bundle.

Abstract: A database system may include a database service and a separate distributed storage service. The database service (or a database engine head node thereof) may be responsible for query parsing, optimization, and execution, transactionality, and consistency, while the storage service may be responsible for generating data pages from redo log records and for durability of those data pages. For example, in response to a write request directed to a particular data page, the database engine head node may generate a redo log record and send it, but not the data page, to a storage service node. The storage service node may store the redo log record and return a write acknowledgement to the database service prior to applying the redo log record. The server node may apply the redo log record and other redo log records to a previously stored version of the data page to create a current version.

Abstract: Techniques for solving quantum circuit mapping problems using reinforcement learning techniques are disclosed. Quantum circuit mapping often requires the use of SWAP gates in order to configure logical quantum computations to be executed using fixed quantum hardware device layouts. A reinforcement learning model takes inputs such as a logical quantum circuit, a physical qubit connectivity graph corresponding to a quantum hardware device, and an initial qubit allocation scheme, and uses such information to schedule quantum gates of the logical quantum circuit for execution using respective physical qubits of the quantum hardware device. A reinforcement learning model that is configured to solve such quantum circuit mapping problems may comprise a neural network that is assisted by a Monte Carlo Tree Search (MCTS) algorithm, wherein the MCTS algorithm guides the neural network towards quantum circuit routing pathways which are more efficient (e.g., require fewer SWAP gates to be scheduled).