Abstract: An example operation may include a method, comprising one or more of: receiving a stop request from a VNFM, retrieving a last peer operational state from a heartbeat history datastore, staying in active state when the last peer operational state is not standby, sending a remaining check points message to a peer VNFCI when the last peer operational state is standby, sending a first heartbeat message to the peer VNFCI with an operational state of active and a desired operational state of shutdown, receiving, at the peer VNFCI, the first heartbeat message; sending a second heartbeat message to the VNFCI, determining an operational state of the VNFCI when the second heartbeat message is received from the peer VNFCI, staying in active state when the operational state in the second heartbeat message is activating, transitioning the VNFCI to a deactivating state, stopping accepting service traffic at the VNFCI, and transitioning to shutdown state at the VNFCI.

Abstract: The present disclosure is generally directed to methods, apparatus, and computer-readable media (transitory and non-transitory) for learning to automatically navigate interactive web documents and/or websites. More particularly, various approaches are presented for training various deep Q network (DQN) agents to perform various tasks associated with reinforcement learning, including hierarchical reinforcement learning, in challenging web navigation environments with sparse rewards and large state and action spaces. These agents include a web navigation agent that can use learned value function(s) to automatically navigate through interactive web documents, as well as a training agent, referred to herein as a “meta-trainer,” that can be trained to generate synthetic training examples. Some approaches described herein may be implemented when expert demonstrations are available. Other approaches described herein may be implemented when expert demonstrations are not available.

Abstract: Systems and methods provide an application programming interface to offer action suggestions to third-party applications using context data associated with the third-party. An example method includes receiving content information and context information from a source mobile application, the content information representing information to be displayed on a mobile device as part of a source mobile application administered by a third party, the context information being information specific to the third party and unavailable to a screen scraper. The method also includes predicting an action based on the content information and the context information, the action representing a deep link for a target mobile application. The method further includes providing the action to the source mobile application with a title and a thumbnail, the source mobile application using the title and thumbnail to display a selectable control that, when selected, causes the mobile device to initiate the action.

Abstract: A disclosed method includes monitoring an audio signal energy level while having a noise suppressor deactivated to conserve battery power, buffering the audio signal in response to a detected increase in the audio energy level, activating and running a voice activity detector on the audio signal in response to the detected increase in the audio energy level and activating and running a noise estimator in response to voice being detected in the audio signal by the voice activity detector. The method may further include activating and running the noise suppressor only if the noise estimator determines that noise suppression is required. The method activates and runs a noise type classifier to determine the noise type based on information received from the noise estimator and selects a noise suppressor algorithm, from a group of available noise suppressor algorithms, where the selected noise suppressor algorithm is the most power consumption efficient.

Abstract: Systems and methods of the present disclosure are directed toward digital signal processing using machine-learned differentiable digital signal processors. For example, embodiments of the present disclosure may include differentiable digital signal processors within the training loop of a machine-learned model (e.g., for gradient-based training). Advantageously, systems and methods of the present disclosure provide high quality signal processing using smaller models than prior systems, thereby reducing energy costs (e.g., storage and/or processing costs) associated with performing digital signal processing.

Abstract: Methods, apparatus, and computer readable media are described related to causing processing of sensor data to be performed in response to determining a request related to an environmental object that is likely captured by the sensor data. Some implementations further relate to determining whether the request is resolvable based on the processing of the sensor data. When it is determined that the request is not resolvable, a prompt is determined and provided as user interface output, where the prompt provides guidance on further input that will enable the request to be resolved. In those implementations, the further input (e.g., additional sensor data and/or the user interface input) received in response to the prompt can then be utilized to resolve the request.

Abstract: The various implementations disclosed herein pertain to methods and systems for detecting a state of monitored objects in a premises. In one aspect, a method is performed at a computer system that is communicably coupled to sensor devices. The computer system receives and processes information of acoustic signals against a database of stored acoustic signatures characterizing predefined acoustic signals generated by electrically non-powered passive tags in response to physical motion of respective monitored objects associated with the passive tags. It receives a first acoustic signal characterized by a first acoustic signature and generated by a first passive tag. In response to the receiving, based on the first acoustic signature and information in the database, it identifies the first monitored object associated with the first acoustic signature, determines a first state of the first monitored object, and provides a notification of the first state to an occupant of the premises.

Abstract: Methods, devices, systems, and means for video integration with a home assistant device are described herein. The home assistant device interacts with a person in a video stream by capturing, using a network-enabled outdoor video camera, a video stream of an outdoor location of a premises at which the person is present and analyzing the person appearing in the captured video stream to determine an identity of the person. Based on determining the identity of the person, the home assistant device announces a presence of the person that is outdoors and outputs instructions to the person.

Abstract: A computer-implemented method for annotating online content with offline interaction data and offline interaction conversion data is implemented by a content annotating computing device in communication with a memory. The method includes receiving a set of content interaction data based on an online interaction between an online user and a first online content item, identifying at least one physical location associated with the online content provider, determining a set of offline interaction conversion data based on the set of content interaction data and a set of offline interaction data, receiving a second online content item associated with the online content provider, and annotating the second online content item based at least partially on the set of offline interaction conversion data and the set of offline interaction data.

Abstract: Methods and apparatus related to determining additional features for a user task entry of a user based on a user habit of the user. For example, one or more aspects of a user task entry of a user may be compared to one or more aspects of a user habit of the user to determine a correlation measure between the task entry and the user habit. If the correlation measure satisfies a threshold, one or more additional features of the user task entry may be determined based on the user habit. For example, the user habit may include one or more trigger indicators that indicate an actual entry or anticipated entry of the user into the user habit, and one or more of the trigger indicators may be utilized to determine task trigger indicators for the user task entry.

Abstract: Processing inputs to electronic list systems. Receiving, from a user device, a user input. Parsing the received user input for a minimum set of product data for querying a shopping service. Upon parsing a minimum set of product data for querying a shopping service, querying at least one shopping service using the parsed data. Receiving, from at least one queried shopping service, at least one product description responsive to the query. Presenting, via the user device, each received at least one product description to the user.

Abstract: Speaker diarization techniques that enable processing of audio data to generate one or more refined versions of the audio data, where each of the refined versions of the audio data isolates one or more utterances of a single respective human speaker. Various implementations generate a refined version of audio data that isolates utterance(s) of a single human speaker by generating a speaker embedding for the single human speaker, and processing the audio data using a trained generative model—and using the speaker embedding in determining activations for hidden layers of the trained generative model during the processing. Output is generated over the trained generative model based on the processing, and the output is the refined version of the audio data.

Abstract: An electronic device tracks its motion in an environment while building a three-dimensional visual representation of the environment that is used to correct drift in the tracked motion. A motion tracking module estimates poses of the electronic device based on feature descriptors corresponding to the visual appearance of spatial features of objects in the environment. A mapping module builds a three-dimensional visual representation of the environment based on a stored plurality of maps, and feature descriptors and estimated device poses received from the motion tracking module. The mapping module provides the three-dimensional visual representation of the environment to a localization module, which identifies correspondences between stored and observed feature descriptors. The localization module performs a loop closure by minimizing the discrepancies between matching feature descriptors to compute a localized pose.

Abstract: To provide ride services within a mapping application in a client computing device without directing the user to a separate ride service application, the mapping application invokes one or several ride service APIs to access ride service data from various ride service providers. For example, the mapping application receives a request for travel directions to a destination and generates multi-modal travel directions which include a route segment where the mode of transportation is a ride service. The mapping application invokes one or several ride service APIs to retrieve a price estimate, estimated wait time, or any other suitable information regarding the ride service route segment. Accordingly, the mapping application provides the multi-modal travel directions to a user including information regarding the ride service route segment.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a sparse mapping of a data hierarchy. In one aspect, a method includes accessing a data hierarchy of elements, receiving a list of identifiers that each identify an element in the data hierarchy; determining, from the data hierarchy and the list of identifiers, a minimum tree that includes each of the identified elements identified by the list of identifiers additional elements that are not identified by the list of identifiers, where the identified elements and the additional elements define a set of a minimum number of elements necessary to form a path in the data hierarchy that connects all of the identified elements identified by the list of identifiers, and where the elements that belong to the minimum tree are a proper subset of elements of the data hierarchy.

Abstract: This patent specification relates to various smart-home systems. Such a system may include a battery-powered smart home device that communicates using a first wireless protocol characterized by relatively low power usage and relatively low data rates. Such a system may further include a smart wall outlet device. The smart wall outlet device may include wireless communication circuitry comprising a first wireless interface and a second wireless interface. The first wireless interface may be configured to communicate with the battery-powered smart home device using the first wireless protocol. The second wireless interface may be configured to serve as a communication bridge between the battery-powered smart home device and a wireless network that uses a second communication protocol characterized by relatively higher power usage and relatively higher data rates.

Abstract: A method includes generating a base model by training with a first dataset of data pairs and generating an adapted model by training the base model on a second dataset of data pairs. The method also includes determining a contrastive score for each data pair of a third dataset of data pairs using the base model and the adapted model. The contrastive score is indicative of a probability of quality of the respective data pair. The method also includes training a target model using the data pairs of the third dataset and the contrastive scores.

Abstract: The technology relates generally to a distributed battery cell charging circuit that allows for battery cells to be positioned in different locations. The distributed battery charging circuit may include a first charging circuit including at least one battery cell, a second charging circuit including at least one battery cell, and a controller configured to control the charging of the at least one battery cell in the first charging circuit independently of controlling the charging of the at least one battery cell in the second charging circuit.

Abstract: Methods, systems, and apparatus, including computer programs encoded on one or more computer storage devices, for training predictive models. Multiple training data records are received that each include an input data portion and an output data portion. A training data type is determined that corresponds to the training data. For example, a training data type can be determined by inputting the output data portions into one or more trained predictive classifiers. In other example, the training data type can be determined by comparison of the output data portions to data formats. Based on the determined training data type, a set of training functions are identified that are compatible with the training data of the determined training data type. The training data and the identified set of training functions are used to train multiple predictive models.

Abstract: Methods, systems, and apparatus, including a method for preventing fraud. In some aspects, a method includes: receiving, from multiple client devices, a measurement data element that includes a respective group member key and a group identifier for a given conversion as a result of displaying a digital component. Each client device uses a threshold encryption scheme to generate, based at least on network data that includes one or more of impression data or conversion data for the conversion, a group key that defines a secret for encrypting the network data and generate, based on data related to the application, the respective group member key that includes a respective share of the secret. In response to determining that at least the threshold number of measurement data elements having the same group identifier have been received, the network data is decrypted using the group member keys in the received measurement data elements.