Abstract: Disclosed herein are computing systems and computer-implemented methods for autonomous vehicle operation. The computing system can include a processor configured to determine a transformation function based on a current position of at least one sensor of a first autonomous vehicle relative to a body of the first autonomous vehicle such that, when the transformation function is applied to images captured by the sensor, the images are adjusted to correspond to an intended position of the sensor relative to the body of the first autonomous vehicle.

Abstract: Techniques for determining a command or intent likely to be subsequently invoked by a user of a system are described. A user inputs a command (either via a spoken utterance or textual input) to a system. The system determines content responsive to the command. The system also determines a second command or corresponding intent likely to be invoked by the user subsequent to the previous command. Such determination may involve analyzing pairs of intents, with each pair being associated with a probability that one intent of the pair will be invoked by a user subsequent to a second intent of the pair. The system then outputs first content responsive to the first command and second content soliciting the user as to whether the system to execute the second command.

Abstract: Aspects of the disclosure relate to transmission of visual content to a display device. A computing platform may receive user information corresponding to a user. The computing platform may determine, based on authentication information in the user information, a profile corresponding to the user, and further determine a visual data file. The computing platform may determine, based on the user information, a display device, among a plurality of display devices. The computing platform may add the visual data file to a playlist associated with the display device. The computing platform may transmit, to the display device, data corresponding to the visual data file.

Abstract: Aspects of the disclosure relate to transmission of visual content to a display device. A computing platform may receive user information corresponding to a user. The computing platform may determine, based on authentication information in the user information, a profile corresponding to the user, and further determine a visual data file. The computing platform may determine, based on the user information, a display device, among a plurality of display devices. The computing platform may add the visual data file to a playlist associated with the display device. The computing platform may transmit, to the display device, data corresponding to the visual data file.

Abstract: A voice user interface (VUI) system use collaborative filtering to expand its own knowledge base. The system is designed to improve the accuracy and performance of the Natural Language Understanding (NLU) processing that underlies VUIs. The system leverages the knowledge of system users to crowdsource new information.

Abstract: Techniques are provided for use in connection with online advertising exchanges. Methods are provided in which information is obtained, including, for each of multiple publisher sites, a set of publisher site tags, and for each of multiple advertisements, a set of advertisement tags. Methods are provided that include, in selection of an advertisement to be served in connection with a serving opportunity, utilizing, as a factor in the selection, matching of publisher site tags, associated with a publisher site associated with the serving opportunity, with advertisement tags of the advertisement.

Abstract: Techniques for determining a command or intent likely to be subsequently invoked by a user of a system are described. A user inputs a command (either via a spoken utterance or textual input) to a system. The system determines content responsive to the command. The system also determines a second command or corresponding intent likely to be invoked by the user subsequent to the previous command. Such determination may involve analyzing pairs of intents, with each pair being associated with a probability that one intent of the pair will be invoked by a user subsequent to a second intent of the pair. The system then outputs first content responsive to the first command and second content soliciting the user as to whether the system to execute the second command.

Abstract: Techniques for determining a command or intent likely to be subsequently invoked by a user of a system are described. A user inputs a command (either via a spoken utterance or textual input) to a system. The system determines content responsive to the command. The system also determines a second command or corresponding intent likely to be invoked by the user subsequent to the previous command. Such determination may involve analyzing pairs of intents, with each pair being associated with a probability that one intent of the pair will be invoked by a user subsequent to a second intent of the pair. The system then outputs first content responsive to the first command and second content soliciting the user as to whether the system to execute the second command.

Abstract: The disclosed systems and methods can provide a closed-loop system that enables network operators to perform streaming analytics for network monitoring applications at scale. The disclosed systems and methods can allow operators to express network monitoring queries as operations over tuples, and can allow them to partition the queries across both switches and a stream processor, and, through iterative refinement, attempt to extract only the traffic that pertains to the query, thus ensuring that the stream processor can scale to satisfy a large number of queries for traffic at very high rates. According to an example method, network monitoring queries are partitioned between components in a network and iteratively refined based on output of the network monitoring query. The network components can include a data plane component (e.g., a switch) and a stream processor component. The network monitoring queries can be refined based on output from the stream processor component.

Abstract: Techniques are provided for use in connection with online advertising exchanges. Methods are provided in which information is obtained, including, for each of multiple publisher sites, a set of publisher site tags, and for each of multiple advertisements, a set of advertisement tags. Methods are provided that include, in selection of an advertisement to be served in connection with a serving opportunity, utilizing, as a factor in the selection, matching of publisher site tags, associated with a publisher site associated with the serving opportunity, with advertisement tags of the advertisement.

Abstract: Embodiments are directed towards dynamically updating, in real-time, an emoticon pool of a user based on a current personal construct of the user. In one embodiment, real-time changes in the current personal construct of a user may result in a real-time update to the emoticon pool of the user, which can allow the user's emoticon pool to dynamically track the user's current personal construct. The current personal construct of a user may be determined by examining user targeted information associated with the user. In some embodiments, the user targeted information may include, but not limited to, static user information, dynamic user information, implicit user information, and/or dynamic external information. In one embodiment, a user's emoticon pool may be dynamically updated by modifying an appearance of one or more base emoticons in a base emoticon pool and/or one or more existing emoticons in the user's emoticon pool.

Abstract: Techniques are provided for use in connection with online advertising exchanges. Methods are provided in which information is obtained, including, for each of multiple publisher sites, a set of publisher site tags, and for each of multiple advertisements, a set of advertisement tags. Methods are provided that include, in selection of an advertisement to be served in connection with a serving opportunity, utilizing, as a factor in the selection, matching of publisher site tags, associated with a publisher site associated with the serving opportunity, with advertisement tags of the advertisement.

Abstract: Embodiments are directed towards dynamically updating, in real-time, an emoticon pool of a user based on a current personal construct of the user. In one embodiment, real-time changes in the current personal construct of a user may result in a real-time update to the emoticon pool of the user, which can allow the user's emoticon pool to dynamically track the user's current personal construct. The current personal construct of a user may be determined by examining user targeted information associated with the user. In some embodiments, the user targeted information may include, but not limited to, static user information, dynamic user information, implicit user information, and/or dynamic external information. In one embodiment, a user's emoticon pool may be dynamically updated by modifying an appearance of one or more base emoticons in a base emoticon pool and/or one or more existing emoticons in the user's emoticon pool.

Abstract: Expanding a user's social connections to include mobile contacts includes: receiving a user's call logs for call/text transactions from a mobile device; deriving usage information from the call logs to determine social pairs; ranking the social pairs based on strength of connectedness; and merging the ranked social pairs with the user's contacts from on-line social networks.

Abstract: Embodiments are directed towards dynamically updating, in real-time, an emoticon pool of a user based on a current personal construct of the user. In one embodiment, real-time changes in the current personal construct of a user may result in a real-time update to the emoticon pool of the user, which can allow the user's emoticon pool to dynamically track the user's current personal construct. The current personal construct of a user may be determined by examining user targeted information associated with the user. In some embodiments, the user targeted information may include, but not limited to, static user information, dynamic user information, implicit user information, and/or dynamic external information. In one embodiment, a user's emoticon pool may be dynamically updated by modifying an appearance of one or more base emoticons in a base emoticon pool and/or one or more existing emoticons in the user's emoticon pool.

Abstract: A first advertisement and a second advertisement are served responsive to a search query. User interactions with the query results, such as with the first advertisement, are detected and used as inputs to selecting at least a portion of the second advertisement. User actions can include a hover, a selection, a button selection, or the like.

Abstract: Techniques are provided for use in connection with online advertising exchanges. Methods are provided in which information is obtained, including, for each of multiple publisher sites, a set of publisher site tags, and for each of multiple advertisements, a set of advertisement tags. Methods are provided that include, in selection of an advertisement to be served in connection with a serving opportunity, utilizing, as a factor in the selection, matching of publisher site tags, associated with a publisher site associated with the serving opportunity, with advertisement tags of the advertisement.

Abstract: Techniques are provided that include utilization of correlation techniques in online advertising. Methods are provided in which, for top-performing articles and advertisements, for example, correlation scores are computed, which provide a quantitative measure relating to performance. An aggregate correlation index may be computed, and individual correlation scores may be determined in relation to the correlation index. In determining positioning of scored articles and advertisements to be displayed on a Web page, preference may be given, as at least one factor in the determination, to position an article adjacent or proximate to an advertisement if they have similar correlation scores.

Abstract: An example of a method includes electronically blending an advertisement into a web page based on physical attributes of the web page and providing the web page with the advertisement.