Abstract: Systems and methods for determining strategy modes for autonomous vehicles are described. An autonomous vehicle may detect aspects of other vehicles and aspects of the environment using one or more sensors. The autonomous vehicle may then determine strategy modes of the other vehicles, and select a strategy mode for its own operation based on the determined strategy modes and an operational goal for the autonomous vehicle. The strategy modes may include an uncoupled strategy mode, a permissive strategy mode, an assistive strategy mode, and a preventative strategy mode. The autonomous vehicle may further determine elements in the environment and topological constraints associated with the environment, and select the strategy mode for its own operation based thereon.

Abstract: Systems and methods for broadcasting strategy modes of autonomous vehicles are described. An autonomous vehicle may include one or more notification devices that broadcast a selected strategy mode for the autonomous vehicle. The selected strategy mode may include an uncoupled strategy mode, a permissive strategy mode, an assistive strategy mode, and a preventative strategy mode. The notification devices may include visual, audio, or signal notification devices. In addition, the autonomous vehicle may broadcast other information associated with the vehicle, such as a position, a velocity, a priority, or an identity of the vehicle.

Abstract: Systems, methods, and computer-readable media are disclosed for auto-scrolling devices and methods. In one embodiment, an example method may include determining a content feed that includes first content and second content for presentation at a user device, and determining a first relevance value for the first content. Example methods may include determining that the first content is relevant to a user account associated with a user of the user device based at least in part on the first relevance value, determining a second relevance value for the second content, and determining that the second content is irrelevant based at least in part on the second relevance value. Example methods may include determining a first scroll speed to scroll through the content feed when the first content is presented by the user device, and determining a second scroll speed to scroll through the content feed when the second content is presented.

Abstract: Disclosed are systems and methods for in-transit or in-flight error correction in state prediction of the vehicle. Natural patterns, such as fields, residential neighborhoods, business areas (e.g., downtown), etc., and/or objects, such as streets, cars, houses, trees, buildings, bridges, roads, parking lots, etc., that are naturally within the environment may be determined and used to correct any error in the aerial vehicle state prediction, while the vehicle is in transit, thereby eliminating or reducing the cumulative error involved in traditional systems.

Abstract: Natural frequencies of vibration of objects, such as aerial vehicles, are identified based on imaging data captured while the objects are subjected to excitation. The imaging data is captured using high-speed cameras, and changes in intensities of pixels corresponding to surfaces of the object are used to determine a spectral diagram of the vibrations from which natural frequencies of vibration (e.g., vibration modes) are determined. The visibility of the vibrating objects is enhanced by providing video images to band-pass filters within small bands around the natural frequencies, and magnifying the vibration based on the amplitudes or phases determined from each of the video images. A stream of the modified video images may be used to determine a mode shape corresponding to the vibration of the objects at or around a natural frequency of vibration.

Abstract: Disclosed are various embodiments for shifting inventory within an inventory network. In a traditional push fulfillment model, items are delivered to customers after orders are placed by customers from a warehouse or fulfillment center through a traditional shipping carrier. Using a pull model as described herein, inventory is shifted throughout an inventory network based upon pull signals generated by sensor networks associated with customers that represent predictions for future orders by customers and the inventory is retrieved by the network node that issues the pull signal.

Abstract: Described herein are a system and methods for providing action recommendations to a user that are likely to result in performance of a high-value action. In some embodiments, the user is compared to one or more other users in order to identify high-value actions for that user. Once at least one high-value action has been identified, a sequence of actions may be generated to include that high-value action using prediction model data that includes probability information. The sequence of actions is then assessed to determine a gateway action within the sequence of actions that is likely to be performed by the user and has a high likelihood of resulting in subsequent performance of the high-value action. The gateway action may then be presented to the user.

Abstract: Systems, methods, and computer-readable media are disclosed for content delivery using persona segments for multiple users. In one embodiment, an example method may include receiving user interaction data from a user device for a session associated with a user account, wherein the user interaction data comprises indications of user interaction of interaction by a user account with a website. A set of keywords may be determined based on the product identifiers. A set of persona segments may be determined based on the keywords. Inter-persona distances between each persona segment of the set of persona segments may be determined. The inter-persona distances may be used to determine that multiple users are using a user account. A persona segment corresponding to a user may be selected. A content may be identified based on the persona segment.

Abstract: Techniques for identifying an emotional response and/or emotional state for a user in response to the user interacting with content may be described herein. For example, a series of facial images of a user may be captured by a user device and transmitted to a service computer that analyzes the images to identify changes in facial animation parameters. An emotional state and/or response may be determined by the service computer by utilizing the change in the facial animation parameters and a facial action coding system. In embodiments, the service computer may edit subsequent content prior to transmitting the content to the user device, utilize a particular communication channel to communicate the content, or identify a workflow that indicates a series of actions and/or content to be presented to the user device based on the determined emotional state/response of the user.

Abstract: Systems, methods, and computer-readable media are disclosed for systems and methods of determining quantitative values representative of user action automaticity. Example methods may include determining a first request for a first user interface from a user device, determining a user identifier associated with the first request, and determining user interaction history data using the user identifier. Example methods may include determining a first selectable option for presentation in a first position at the first user interface using the user interaction history, determining a second selectable option for presentation in a second position at the first user interface, generating the first user interface, and sending the first user interface to the user device.

Abstract: Systems and methods to calibrate an imaging sensor may include an enclosure with a controlled environment, a light source illuminating the environment, and a particle source emitting desired particles at desired concentrations into the environment. An imaging sensor, which may be associated with an aerial vehicle, may be placed within the enclosure, and the imaging sensor may capture imaging data within the controlled environment. In addition, the imaging data may be processed to determine current spectral characteristics of the imaging sensor. Based on the environment, light, and particle properties within the environment, the imaging sensor may be calibrated to exhibit nominal or desired spectral characteristics.

Abstract: Techniques for tracking deliveries of items and verifying route adherence by a delivery person may be described herein. For example, first information may be maintained that identifies an association of a virtual asset with a machine-readable identifier. Further, second information may be generated and maintained that identifies an association of the machine-readable identifier with an item. Upon receiving an order for an item, delivery instructions may be generated and provided to an entity for delivering the item to a location. The delivery to the location may be verified based at least in part on geo-location information provided by a user device interacting with the machine-readable identifier of the item. In embodiments, the virtual asset associated with the machine-readable identifier may be associated with a user profile corresponding to the delivery location instead of with the machine-readable identifier.

Abstract: Identifiers or references to supplemental information or content regarding images may be steganographically encoded into the images. The identifiers or references may be encoded into least significant bits or less significant bits of pixels within the image that may be selected on any basis. The identifiers or references may include alphanumeric characters, bar codes, symbols or other features. When an image is captured of an image having one or more identifiers or references steganographically encoded therein, the identifiers or references may be interpreted, and the supplemental information or content may be accessed and displayed on a computer display. In some embodiments, the supplemental information or content may identify and relate to a commercial product expressed in an image, and may include a link to one or more pages or functions for purchasing the commercial product.

Abstract: Systems, methods, and computer-readable media are disclosed for systems and methods for automatic execution of actions responsive to queries. Example methods include determining a set of seed words for input at a first search engine, determining a first autocomplete result associated with a first word at the first search engine, and determining a second autocomplete result associated with the first autocomplete result at the first search engine. Example methods include determining a set of user interaction histories that comprises queries input at a second search engine, and determining that a first user interaction history includes a query with a portion of the second autocomplete result. Example methods include determining a user action performed after the query, and associating the user action with the query in a database, such that subsequent inputs of the query cause the first user action to automatically occur.

Abstract: A content provider may cause a client device of a user to output a personalized puzzle in response to receiving a request from the client device to access electronic content of the content provider. The puzzle may include a theme that corresponds to a determined predilection of the user, and/or the puzzle may be a type of puzzle that corresponds to the user's predilection. The client device may also output, with the puzzle, an incentive for completing (e.g., solving) the puzzle Upon receiving data indicating that the user has completed his/her personalized puzzle, the content provider may provide the reward to the user.

Abstract: In a machine learning environment, missing values can be imputed based upon an expectation maximization style approach. A system partitions a dataset, and uses a first partition as a training subset and a second partition as a verification subset. The training subset is used to train a machine learning model, which is then used to impute missing values in the second subset. The subsets may be swapped and the process iterates to predict missing values in the dataset with a high degree of accuracy, thereby improving both the accuracy of the machine learning model and the accuracy of the imputed values. The noise in the value prediction is reduced through a linear regression setting to account for heterogeneity in the dataset.

Abstract: Techniques for managing notifications may be described. In an example, the notifications may relate to an item and may be provided to a user device. An active device may be associated with the item. The active device may store a token for communication with a local area network associated with a location. Based on the communication, a determination may be made that the item may be in proximity to the location. Corresponding notifications may be sent to the user device.

Abstract: Navigable path networks are generated or modified based on attributes of tasks to be performed by autonomous vehicles traveling thereon, attributes of such vehicles, or attributes of the environments in which such networks are provided. The networks include traditional and non-traditional transportation features, and are defined based on prior travel within the environments, including information gathered by such vehicles during such prior travel. The autonomous vehicles are robotic, self-powered units having storage compartments for carrying objects between points of the networks. An optimal route within a navigable path network is selected based on attributes of an autonomous vehicle, a task to be performed by the autonomous vehicle, or the various paths within the network. A navigable path network is updated based on information subsequently learned regarding the environment, including information captured by autonomous vehicles traveling on paths of the network.

Abstract: Techniques for maintaining information that identifies an association of a virtual asset to an item, updating the association of the virtual asset to a user profile, and generating a recommendation for another item based on the updated association may be described herein. For example, data may be received in response to a user device interacting with a machine-readable identifier of an item. An association of a virtual asset may be updated to reflect that the virtual asset is associated with a user profile instead of the item in response to receiving the data. In some embodiments, a recommendation may be generated and provided to the user device that identifies other vendors or items based at least in part on the updated user profile.

Abstract: A vibrometric signature for a vehicle, or a set of one or more frequencies where vibration of the vehicle is naturally observed in the presence of excitation, may be generated and used to make one or more determinations regarding the integrity or suitability of the vehicle for one or more missions. When the vehicle is subjected to excitation over a range of frequencies, images of the vehicle are captured, and power levels of vibrations of the vehicle are calculated based on the images. A vibrometric signature is generated based on the power levels of the vibrations, and compared to vibrometric signatures previously generated for the aerial vehicle, or to vibrometric signatures associated with one or more other vehicles, or anomalies experienced by such other vehicles, to determine whether the vehicle may be cleared for the performance of one or more missions, or whether maintenance or inspections are required.