Abstract: Examples are disclosed that relate to providing image data to a user in a defined space of a surrounding environment from a perspective of the user. One example provides a computing system, comprising a logic subsystem, and a storage subsystem comprising instructions executable by the logic subsystem to obtain information regarding a pose of a user within a defined space, based upon the pose of the user, determine a portion of an environment surrounding the defined space toward which the user is looking, obtain image data representing the portion of the environment from a perspective of the user, and provide the image data for display via a display device within the defined space.

Abstract: A system and method and for retrieving one or more one or more multimodal assets includes receiving a search query for searching for one or more multimodal assets from among a plurality of candidate multimodal assets, encoding the search query into one or more query embedding representations via a trained query representation machine-learning (ML) model, comparing, via a matching unit, the one or more query embedding representations to a plurality of multimodal tensor representations, each of the plurality of multimodal tensor representations being a representation of one of the plurality of candidate multimodal assets, and identifying, based on the comparison, at least one of the plurality of the candidate multimodal assets as a search result for the search query, and providing the at least one of the plurality of the candidate multimodal assets for display as the search result.

Abstract: Techniques are described for low-latency streaming of media content using a lossless protocol. For example, a media stream, comprising encoded video data, can be streamed to a plurality of streaming clients via the lossless protocol. The plurality of streaming clients can be monitored to determine whether any of them have fallen behind in streaming the media stream. When a streaming client falls behind, a portion of video data to be streamed to the streaming client can be selectively dropped based on scalability information and/or long term reference (LTR) frame information. The low-latency streaming can be performed without using per-client quality feedback from the plurality of streaming clients. When streaming using a semi-lossy protocol, a plurality of delivery modes can be used, where each delivery mode is for a different type of encoded video data and provides a different level of reliability.

Abstract: A method to control an output component of an electronic system comprises (a) receiving a transmission from an input component of the electronic system, the transmission including a time stamp and at least one input signal; (b) storing content of the transmission including the time stamp and the at least one input signal; (c) selecting one of a plurality of noise-filtered signals based on the time stamp and on a reference time index, the selected one of the plurality of noise-filtered signals having a greatest signal-to-noise ratio among the noise-filtered signals defined at the reference time index; and (d) controlling an output component of the electronic system based in part on the selected noise-filtered signal.

Abstract: Various embodiments of a modular unit for a topologic qubit and of scalable quantum computing architectures using such modular units are disclosed herein. For example, one example embodiment is a modular unit for a topological qubit comprising 6 Majorana zero modes (MZMs) on a mesoscopic superconducting island. These units can provide the computational MZMs with protection from quasiparticle poisoning. Several possible realizations of these modular units are described herein. Also disclosed herein are example designs for scalable quantum computing architectures comprising the modular units together with gates and reference arms (e.g., quantum dots, Majorana wires, etc.) configured to enable joint parity measurements to be performed for various combinations of two or four MZMs associated with one or two modular units, as well as other operations on the states of MZMs.

Abstract: A virtual meeting server and method to provide automatic status notifications of screensharing status between a presenter device sharing a screen with at least one participant device during a screensharing session, including a channel module to establish and maintain communication channels between the presenter and participant devices for transmitting signals to set up the screensharing session and data, including content frames, used in the screensharing session, a detection module to detect an acknowledgement response that the participant device has received the content, an analysis engine to analyze and determine a status of the screensharing session based on analyzing the acknowledgement response, and a notification module to send a status notification to the presenter device and the participant device, wherein the status notification indicates whether connection for the screensharing session has been established and is being maintained between the presenter device and the participant device.

Abstract: A personalized natural language processing system tokenizes a plurality of sets of raw text data to generate a plurality of sets of tokenized text data for the plurality of users, respectively. The tokenized text data includes a sequence of tokens corresponding to the raw text data, the tokens at least identifying distinct words or portions of words in the raw text. The system appends predetermined user-specific tokens to the sets of tokenized text data from the users, respectively. Each predetermined user-specific token corresponds to one of the users. The system processes the sets of tokenized text data using the NLP model in accordance with the appended predetermined user-specific tokens to predict a personalized classification for the sets of tokenized text data from each of the users, and outputs the personalized classifications of the tokenized text data for each of the users.

Abstract: A centralized skills management server, a computer-readable storage medium, and a computer-implemented method for skills inference are described herein. The method includes executing a web-based application on a remote computing system operated by a user associated with a tenant and extracting skills-related terms associated with the execution of the web-based application. The method includes interfacing with the global skills graph via an API and importing standardized skill tags relating to the extracted skills-related terms.

Abstract: According to examples, an apparatus may include a processor and a memory on which are stored machine-readable instructions that, when executed by the processor, may cause the processor to receive event data to be placed in a plurality of queues. The processor may determine logical partitions for the received event data and may store the event data in the plurality of queues based on the determined logical partitions. A first queue of the plurality of queues may be mapped to a first logical partition range and a second queue of the plurality of queues may be mapped to a second logical partition range. Based on respective loads at the plurality of queues, the processor may update a mapping of the determined logical partitions to the plurality of queues by transferring logical partitions between the first queue and the second queue.

Abstract: A system and method to provide automatic status notifications of screensharing status to meeting participant devices participating in a screensharing session in a virtual meeting via a virtual meeting server, to, after an activation of the screen sharing session, determine if content frames for the screensharing session sent from a presenter device have reached other receiving meeting participant devices, determine if the received content frames meet predetermined threshold requirements, determine if an acknowledgment response has been sent to the virtual meeting server from a receiving device confirming receipt of the content frames, and generate a status notification to indicate on a user interface of the presenter device whether the content frames have been received by the receiving devices and meet the predetermined threshold requirements or fail to meet the predetermined threshold requirements.

Abstract: Methods and systems to normalize an input table having a plurality of input table columns with a normalized table having a plurality of normalized table columns are disclosed. For each normalized column identifier associated with a normalized column of the normalized table, a compatibility score is computed for the normalized column identifier and each input column identifier associated with an input column of the input column table to provide set of compatibility scores associated with each normalized column identifier and input column identifier pair. A combinatorial optimization is applied to determine a match for each normalized column identifier with an input column identifier. Data associated with an input column of the input column identifier is mapped to the normalized column of the normalized column identifier matched with the input column identifier.

Abstract: A centralized skills management server, a computer-readable storage medium, and a computer-implemented method for standardized skill tagging are described herein. The method includes accessing, via a processor of the computing system, a property graph including data objects corresponding to a tenant and extracting skills-related terms from object metadata corresponding to the data objects within the property graph. The method also includes interfacing with a global skills graph via an API and importing standardized skill tags from the global skills graph, where the imported standardized skill tags relate to the extracted skills-related terms from the object metadata. The method further includes incorporating the standardized skill tags into the object metadata for the corresponding data objects within the property graph.

Abstract: Techniques are described for providing real-time audio and/or video feedback during a conference call. Audio feedback can be provided during a conference call in response to a participant unmuting the participant's microphone. For example, a feedback period can be initiated upon receiving an unmute indication. During the feedback period, the participant's microphone audio can be sent, in real-time, back to the participant for playback in addition to sending to the other participants. After the feedback period is over, the participant can be removed from receiving their microphone audio. Video feedback can be provided during a conference call in response to screen sharing. For example, during a feedback period, the participant can receive, in real-time, a screen content thumbnail of their shared screen content. After the feedback period is over, the screen content thumbnail can stop being sent.

Abstract: The description relates to hinged devices. One example relates to a device that has a first portion hingedly secured along a spine and a second portion hingedly secured along the spine. The example can also include a first spine cover secured relative to the spine and extending into the first portion and a second spine cover secured relative to the spine and extending into the second portion. An extent to which the first spine cover extends into the first portion and the second spine cover extends into the second portion is controlled by an angular orientation of the first and second portions.

Abstract: Aspects of the present disclosure relate to video stream refinement for a dynamic scene. In examples, a system is provided that includes at least one processor, and memory storing instructions that, when executed by the at least one processor, causes the system to perform a set of operations. The set of operations include receiving an input video stream, identifying, within the input video stream, a frame portion containing features of interest, enlarging the frame portion containing the features of interest, enhancing the frame portion of the input video stream to increase fidelity within the frame portion, and displaying the enhanced frame portion.

Abstract: Examples are disclosed that relate to motion blur corrections for time-of-flight (ToF) depth imaging. One example provides a depth camera comprising a ToF image sensor, a logic machine, and a storage machine storing instructions executable by the logic machine to receive depth image data from the ToF image sensor, the depth image data comprising phase data and active brightness (AB) data, determine a first two-dimensional (2D) AB image corresponding to a first modulation frequency, and determine a second 2D AB image corresponding to a second modulation frequency. The instructions are further executable to determine a 2D translation based upon a comparison between the first 2D AB image and the second 2D AB image, determine corrected phase data based on the 2D translation to form corrected phase data, perform phase unwrapping on the corrected phase data to obtain a three-dimensional (3D) depth image, and output the 3D depth image.

Abstract: Examples of the present disclosure describe systems and methods for preventing illicit data transfer and storage. In aspects, a computing platform may receive a data request from a caller system, device, or service. The computing platform may identify data items/properties associated with the data request and retrieve one or more rules relevant to the caller and/or caller location. The retrieved rule(s) may be used to evaluate the data item(s) such that data items, data item content, and/or data item properties that are prohibited by the retrieved rule(s) from being manipulated (e.g., accessed, transferred, stored) are removed from the identified data item(s). Based on the evaluation of the identified data item(s), one or more relevant status codes may be set. The computing platform may then manipulate the identified data item(s) in accordance with the data request and provide a processing response to the caller.

Abstract: Aspects of systems and methods for providing contextual and event driven insights are provided. The system monitors information about the users and their conversations. Upon receiving a natural language request for information for a topic, the system utilizes a model to extract one or more topics from the request. The system utilizes the topic to query a resource for candidate users with knowledge about the topic. The system then queries a resource to identify candidate content items associated with the topic and the candidate users. Thereafter, the system refines the candidate users and the candidate content items to identify relevant users and content items that are meaningful to the user.

Abstract: Provided is a method, performed by a web browser, for identifying and displaying content, the method including receiving a webpage including text, comparing the text to titles of media content in a media content database, based on the comparison, identifying a subset of text as a match to a title of a matching media content, based on identifying the subset of text, modifying a display of the webpage by inserting a visual indicator into the webpage in proximity to the subset of text, or modifying the display of the subset of text, detecting an interaction of a user-controlled input with the inserted visual indicator or the modified display of the subset of text, and displaying a popup interface over a portion of the webpage, the popup interface displaying additional information about the matching media content and a selectable user-interface element to play the matching media content.

Abstract: An automatic speech recognition (ASR) system recognizes speech expressed in different languages. The ASR system includes a language-agnostic encoding component and prediction component. A language-specific joint analysis system generates first-language probabilities for symbols of a first language and second-language probabilities for symbols of a second language, based on outputs generated by the encoding component and the prediction component. The ASR system then modifies the probabilities produced by the joint analysis system by language-specific weighting information that, in turn, is produced by an attention system. This yields modified first-language probabilities and modified second-language probabilities. Finally, the ASR system predicts an updated instance of label information based on the modified first-language probabilities and the modified second-language probabilities. The ASR system can be successfully applied to recognize an utterance that combines words in two or more languages.

Abstract: This document describes techniques and apparatuses for managing an immersive interface in a multi-application immersive environment. In some embodiments, these techniques and apparatuses enable a user to alter sizes and/or a layout of multiple immersive interfaces with as little as one selection.

Abstract: Aspects of the technology described herein provide a collaborative browsing experience in which real-time browsing activity and saved browsing activity of session collaborators in a collaborative browsing session are shared with the collaborators. A collaborative session may be initiated, which may create a tab group associated with the session and linked to a collaborator. Other collaborators may be invited to join the session, and additional tab groups for each collaborator may be created. The tab groups of the collaborators may be included in a collective tab group, which may be updated in real-time with changes made by any of the collaborators. For example, client changes may be handled locally and communicated to a service to which each client is connected. The service may sequence and broadcast the ordered changes to the clients, which may each implement the changes according to the sequence to synchronize a shared state amongst clients.

Abstract: Systems and methods are provided for determining faces and bodies of people in an image by adaptively scaling images and by iteratively using a deep neural network for inferencing. A camera captures an image including faces and bodies of people. A face/body determiner determines faces and bodies of people appearing in the image by resizing the image into a predetermined pixel dimension as input to the deep neural network. A region cropper determines a crop region associated with a low level of confidence in detecting faces and bodies that are too small to determine with an acceptable level of confidence. The region cropper resizes the crop region into the predetermined pixel dimension as input to the deep neural network. The face and body determiner determines other faces and bodies appearing in the resized crop region. An aggregator aggregates locations of the determined faces and bodies in the image.

Abstract: A computing device is provided. The computing device may include a first camera configured to capture a primary image sequence of a scene, and a second camera configured to substantially concurrently capture a secondary image sequence of the scene. The computing device may include a processor configured to execute instructions stored in memory to obtain a depth map for an image of the primary image sequence, and based on the depth map, generate a reference frame of background pixels from the primary image sequence. The processor may be configured to execute the instructions to detect an intrusive object in the secondary image sequence, replace the intrusive object in the primary image sequence with corresponding background pixels of the reference frame to generate a modified primary image sequence, and output on a display at least one image of the modified primary image sequence.

Abstract: This document relates to reinforcement learning. One example includes performing two or more training iterations to update a policy. Individual training iterations can be performed by a training process executing on a training computing device. The training iterations can include obtaining experiences representing reactions of an environment to actions taken by a plurality of remote agent processes according to the policy. The remote agent processes can execute the policy on remote agent computing devices and the experiences can be obtained from the remote agent computing devices over a network. The training iterations can also include updating the policy based on the reactions of the environment to obtain an updated policy and distributing the updated policy over the network to the plurality of remote agent processes.

Abstract: A method for source attribution comprises receiving measurements of a chemical species at a spatially distributed sensor array for a given set of spatially positioned emission sources in a physical environment using a dispersion model. Based on the received measurements, a concentration field is mapped from the emission sources to the sensor array using a forward operator. For each emission source, a likelihood data set is evaluated at least by fitting an emission rate of the chemical species using a regression model based on the mapped concentration field and real-world, runtime measurements from the sensor array. A posterior data set is evaluated based at least on the evaluated likelihood data set and historical data for the physical environment. For each sensor of the sensor array, estimated emission rates and contribution rankings for emission sources are determined and output based on the evaluation of the posterior data set.

Abstract: A method and system for transforming an input image via a plurality of image transformation stylizers includes receiving the input image; providing the input image, information about the plurality of image transformation stylizers and at least one of user data, history data, and contextual data to a trained machine-learning (ML) model for selecting a subset of the plurality of image transformation stylizers; receiving as an output from the ML model the subset of image transformation stylizers; executing the subset of the image transformation stylizers on the input image to generate a plurality of transformed output images; ranking the plurality of transformed output images based on at least one of the input image, the user data, the history data, and the contextual data; and providing the ranked plurality of transformed output images for display.

Abstract: A document re-finding system generates embeddings for concept clips provided by a user, the concept clips defining a concept for searching for content of interest to the user in a plurality of documents previously seen by the user. The re-finding system determines semantic relationships between the concept and document clips, related to respective document among the plurality of documents, based on the concept embedding and embeddings generated for the document clips. A graphical user interface depicting the semantic relationships is rendered to the user and is operable to enable re-finding a document, among the plurality of documents, having the content of interest to the user.

Abstract: Aspects of the present disclosure relate to systems and methods for dynamically adjusting queries based on data protection configurations. In one example, a request for data is received from an application. The data is indexed in at least one of a first index database, a second index database and a third index database. A first key notification indicating a protection status of a personal device is received. A query is sent to each of the first index database, the second index database and the third index database for the requested data based on the protection status of the personal device. The requested data is loaded into the application. A second key notification indicating a change in the protection status of the personal device is received. At least some of the loaded data is removed from the application based on the change in the protection status of the personal device.

Abstract: A method for predicting the pose of an articulated object includes receiving spatial information for n joints of the articulated object. The spatial information for the n joints is passed to a machine learning model previously trained to receive spatial information for n+m joints as input, wherein m>=1. From the machine learning model, a pose prediction for the articulated object is received as output based at least on the spatial information for the n joints, and without spatial information for the m joints.

Abstract: A broad-spectrum laser for use in a MEMS laser scanning display device is provided. In one example, the broad-spectrum laser includes a laser diode emitter with plural quantum wells each having a different spectral peak. In another example, the broad-spectrum laser includes a laser diode emitter with a tunable absorber to achieve a broadened emissions spectrum. In another example, the broad-spectrum laser includes a laser diode emitter array having plural individual emitters with different spectral peaks.

Abstract: A computer system is provided, including a processor and associated memory storing instructions that when executed cause the processor to implement a plurality of artificial intelligence (AI) models. Each AI model is configured to receive, as input, time series data and to output a model-specific time series forecast including a respective predicted value for each of a plurality of future time steps. The processor is further configured to implement a model selection neural network configured to select a predicted most accurate AI model from among the plurality of AI models for each of the plurality of future time steps. The processor is further configured to implement a blended output generator configured to output a model-blended time series forecast including the respective predicted value computed by the predicted most accurate AI model selected for each of the plurality of future time steps.

Abstract: This document relates to the compressing and/or decompressing of spatial data. A block of spatial data can be compressed by iteratively performing compression iterations on portions of the block and splitting the portions into further portions until compression is completed. The compressed data can include first encoded values indicating whether matches were obtained for comparisons to test values during the compression iterations. The compressed data can also include second encoded values reflecting results of one or more modifications performed on the test values during the compression iterations.

Abstract: The disclosed technology is generally directed to the identification of unobservable nodes. In one example of the technology, a set of monitored machines are monitored. Each of the monitored machines includes associated monitoring logic. Monitoring the set of monitored machines includes receiving the machine information from the monitoring logic. A total inventory list of the set of monitored machines is generated in a manner that is independent of the monitoring of the monitored machines. An agent inventory list is generated such that the agent inventory list is a list of machines that, based on the monitoring of the monitored machine, are determined to have been running monitoring logic during a particular time period. Unobservable machine information that is associated with a set of unobservable machines is determined based on a comparison of the total inventory list with the agent inventory list.

Abstract: A data processing system implements obtaining textual content in a first language from a first client device and segmenting the textual content into a plurality of first tokens. The system also implements translating the first tokens from the first language to a second language using a bilingual dictionary, extracting features information from the second tokens to create a features vector, providing the feature vector to a first natural language processing model trained to analyze textual input in the second language and to output contextual information indicating one or more topics or subject matter of the first textual content, and providing the contextual information to a first machine learning model configured to analyze the contextual information and to identify one or more content items predicted to be relevant to the contextual information. The system further implements providing the information identifying the one or more content items to the first client device.

Abstract: Examples are disclosed herein relating to reducing strain in a resonant scanning mirror system. One example provides a thin film piezoelectric-actuated resonant scanning mirror system, comprising a body comprising an anchor portion, a scanning mirror portion, a piezoelectric film support portion, a transmission beam extending from the piezoelectric thin film support portion, and a torsion beam extending between the scanning mirror portion and the transmission beam, and a piezoelectric film formed on the piezoelectric film support portion, the piezoelectric film support portion comprising an area of a surface of the body in which a stress on the piezoelectric film does not exceed a yield stress of the piezoelectric film during oscillation of the scanning mirror portion.

Abstract: A method of and system for synchronous communications between a sender and recipient(s) of a message containing a question include receiving over a network an indication that the message was sent from the sender to the recipients via a user interface of a first application or a first service and a response to the question was not received within a predetermined time, monitoring a digital location of the sender and the recipients, the digital location being an electronic application or service being currently utilized, determining whether the sender and at least one of the recipients are concurrently using a second application or a second service, and upon determining that the sender and at least one of the recipients are concurrently using the second application or the second service, sending a notification signal over the network to the sender or one of the one or more recipients to notify the sender or the recipient via a user interface element displayed on a client device that the sender and at least one o

Abstract: A head-mounted display system includes a left display assembly configured to provide left-side display light. A first left inertial measurement unit (IMU) is configured to measure an orientation of the left display assembly. A right display assembly is configured to provide right-side display light. A first right IMU is configured to measure an orientation of the right display assembly. A communication interface is configured to receive a left-side orientation of a head-tracking system as measured by a second left IMU, and a right-side orientation of a head-tracking system as measured by a second right IMU. A logic machine is configured to assess an alignment of the head-mounted display system based at least in part on the orientation of the left display assembly, the orientation of the right display assembly, the left-side orientation of the head-tracking system, and the right-side orientation of the head-tracking system.

Abstract: Examples of the present disclosure describe systems and methods for determining relationships between content items to create a visualization associated with the various content items. The visualization may provide information regarding what various individuals in a group, team, or organization have been working on (e.g., content, documents, projects).

Abstract: A semiconductor device is fabricated by: forming a shadow wall on a substrate; subsequently growing a nanowire of semiconductor material on the substrate; and directionally depositing a layer of a further material on the nanowire from a direction selected such that the shadow wall casts a shadow on the nanowire, the shadow being a region in which the further material is not deposited. The nanowire is vertically orientated relative to the substrate. The shadow wall comprises a base portion and a bridge portion. The bridge portion overhangs the substrate and is supported by the base portion. Patterning of the further material may be achieved without the use of etching, thereby avoiding damage to the semiconductor. Also provided is a semiconductor-superconductor hybrid device; a quantum computing device comprising the semiconductor-superconductor hybrid device; and a shadow wall for controlling directional deposition of a material.

Abstract: A head-mounted display system includes a left display assembly configured to provide left-side display light and left-side test light. A left waveguide incouples the left-side display light and outcouples the left-side display light for viewing. A left optical sensor is positioned to measure the left-side test light. A left inertial measurement unit (IMU) is configured to measure an orientation of the left display assembly. A right display assembly is configured to provide right-side display light and right-side test light. A right waveguide incouples the right-side display light and outcouples the right-side display light for viewing. A right optical sensor is positioned to right the right-side test light. A right IMU is configured to measure an orientation of the left display assembly. A logic machine is configured to assess a stereo alignment for the left- and right-side display light.

Abstract: A method for providing graph data is described. A request for graph data based on a data graph is received, the data graph having i) nodes representing entities associated with an enterprise organization, and ii) edges between nodes representing relationships among the entities. A search embedding corresponding to the request is generated. Embeddings from a set of embeddings that are adjacent to the search embedding are identified, wherein the set of embeddings represent the data graph. Graph data corresponding to the identified embeddings is provided in response to the request.

Abstract: Examples of the present disclosure describe systems and methods for implementing a success rate SLI solution for service pipelines. In aspects, a metrics service may detect that a number of payloads relating to one or more activities or service requests have been received at one or more services of a service pipeline. For each payload processed by a service of the service pipeline, the metrics service may determine a set of payload processing metrics for the service. The set of payload processing metrics for the service may be applied to the payload. The payload processing metrics for each service may be aggregated and used to calculate a success rate for payloads processed using the service pipeline. Based on the success rate, an SLI may be evaluated and/or an action associated with the activity/service request may be performed.

Abstract: Systems and methods are provided for determining an optical bypass for an inter-regional wide area network (WAN) for regions of server facilities of a cloud service provider. In particular, the optical bypass connects non-adjacent regional server centers of the WAN by eliminating needs of data conversions at intermediate regional server centers. The determining the optical bypass includes receiving a WAN topology data, capacity and demand information about the WAN. The determining includes an objective function to maximize a number of network resources to free up by determining a revised data flow and bandwidth allocations by introducing the optical bypass in the WAN. The disclosed technology transmits the determined data traffic flow and resource allocation information of the optical bypass, causing a network traffic enforcers to reconfigure the WAN.

Abstract: Methods, systems, and computer programs are presented for automatically shutting down idle virtual machines (VMs). One method includes an operation for determining a number of active connections of a VM. The method further includes checking disconnect rules to determine automatic shutting down of the VM. The disconnect rules are based on the number of active connections of the VM, and the disconnect rules define criteria for determining when to shut down the VM automatically. The method further includes, based on determining to automatically shut down the VM, sending a message indicating the VM is inactive to a VM manager program executing at a cloud-service provider. The VM manager program is configured to shut down the VM in response to receiving the message indicating the VM is inactive.

Abstract: Innovations in adaptive encoding and decoding for units of a video sequence can improve coding efficiency. For example, some of the innovations relate to encoding/decoding that includes adaptive switching of color spaces between units within a video sequence. Other innovations relate encoding/decoding that includes adaptive switching of color sampling rates between units within a video sequence. Still other innovations relate encoding/decoding that includes adaptive switching of bit depths between units within a video sequence.

Abstract: Examples are disclosed that relate to handling a circularly polarized signal via a plurality of linearly polarized antennas. One example provides a mobile device comprising an inertial measurement unit (IMU) and an antenna system configured for communication using a circularly polarized signal. The antenna system comprises a first linearly polarized antenna, a second linearly polarized antenna, a third linearly polarized antenna, and a processing stage. The processing stage is configured to adjust, based at least in part on data from the IMU, one or more of a phase or a gain of a signal on each of the first, second, and third linearly polarized antennas to direct a beam of the antenna system toward a direction of the circularly polarized signal.

Abstract: The description relates to artificial intelligence assisted wearables, such as backpacks. An example backpack may include sensors, such as a microphone and a camera. The backpack may receive a contextual voice command from a user. The contextual voice command may include a non-explicit reference to an object in an environment. The backpack may use the sensors to sense the environment, use an artificial intelligence engine to identify the object in the environment, and use a digital assistant to perform a contextual task in response to the contextual voice command. The contextual task may relate to the object in the environment. The backpack may output a response to the contextual voice command to the user.

Abstract: Methods, systems, and computer programs are presented for redirecting executing of a virtual machine (VM) program to a client device. One method is performed by a server executing the VM. The method includes an operation for receiving an input from a remote desktop application of a client device to execute a program at the VM, and for checking redirect logic to determine execution of the program on the client device. The redirect logic comprises at least one rule to redirect execution of the program to the client device instead of executing the program at the VM. Further, the method includes an operation for, based on determining to execute the program on the client device, send to the client device a request for executing the program at the client device. The client device is configured to execute the program in response to the request.

Abstract: Knowledge graphs can greatly improve the quality of content recommendation systems. There is a broad variety of knowledge graphs in the domain including clicked user-ad graphs, clicked query-ad graphs, keyword-display URL graphs etc. A hierarchical Transformer model learns entity embeddings in knowledge graphs. The model consists of two different Transformer blocks where the bottom block generates relation-dependent embeddings for the source entity and its neighbors, and the top block aggregates the outputs from the bottom block to produce the target entity embedding. To balance the information from contextual entities and the source entity itself, a masked entity model (MEM) task is combined with a link prediction task in model training.