Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a document classification neural network. One of the methods includes training an autoencoder neural network to autoencode input documents, wherein the autoencoder neural network comprises the one or more LSTM neural network layers and an autoencoder output layer, and wherein training the autoencoder neural network comprises determining pre-trained values of the parameters of the one or more LSTM neural network layers from initial values of the parameters of the one or more LSTM neural network layers; and training the document classification neural network on a plurality of training documents to determine trained values of the parameters of the one or more LSTM neural network layers from the pre-trained values of the parameters of the one or more LSTM neural network layers.

Abstract: Provided herein are devices, systems, and methods for a three-dimensional registering, tracking, and/or guiding an object of interest, such a body part, a surgical tool, or an implant, during a surgical procedure. Such devices, systems, and methods may offer minimally invasive, high precision registering, tracking, and/or guiding of the object of interest using a patterned light beam and data processing using artificial intelligence. The methods, devices, and systems disclosed herein may be compatible with a simple marker or markers placed by a minimally invasive method on the object of interest.

Abstract: Provided is an electronic device including: a housing; a camera module oriented in a first direction inside the housing; a camera window arranged in front of the camera module in the first direction; a cover including an opening into which the camera window is insertable, the cover coupled to the housing; and a window frame supporting the camera window and mounted on the cover. The window frame includes: a frame body including a support portion arranged in the opening and supporting the camera window, and a mounting portion extending from the support portion in a second direction that is perpendicular to the first direction and overlapped by the cover in the first direction; a retention member retaining a first gap between the mounting portion and the cover in the first direction; and a waterproof body configured to be elastically deformed while arranged between the mounting portion and the cover.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating speech from text. One of the systems includes one or more computers and one or more storage devices storing instructions that when executed by one or more computers cause the one or more computers to implement: a sequence-to-sequence recurrent neural network configured to: receive a sequence of characters in a particular natural language, and process the sequence of characters to generate a spectrogram of a verbal utterance of the sequence of characters in the particular natural language; and a subsystem configured to: receive the sequence of characters in the particular natural language, and provide the sequence of characters as input to the sequence-to-sequence recurrent neural network to obtain as output the spectrogram of the verbal utterance of the sequence of characters in the particular natural language.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating document vector representations. One of the methods includes obtaining a new document; and determining a vector representation for the new document using a trained neural network system, wherein the trained neural network system has been trained to receive an input document and a sequence of words from the input document and to generate a respective word score for each word in a set of words, wherein each of the respective word scores represents a predicted likelihood that the corresponding word follows a last word in the sequence in the input document, and wherein determining the vector representation for the new document using the trained neural network system comprises iteratively providing each of the plurality of sequences of words to the trained neural network system to determine the vector representation for the new document using gradient descent.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a computer chip placement. One of the methods includes obtaining netlist data for a computer chip; and generating a computer chip placement, comprising placing a respective macro node at each time step in a sequence comprising a plurality of time steps, the placing comprising, for each time step: generating an input representation for the time step; processing the input representation using a node placement neural network having a plurality of network parameters, wherein the node placement neural network is configured to process the input representation in accordance with current values of the network parameters to generate a score distribution over a plurality of positions on the surface of the computer chip; and assigning the macro node to be placed at the time step to a position from the plurality of positions using the score distribution.

Abstract: A device holder assembly that includes a body and a housing coupled to the body. A drive assembly is rotatably coupled to the housing. A leg assembly that includes a first leg and a second leg is coupled to the housing. A first rack gear is coupled to the first leg and the drive assembly. A second rack gear is coupled to the second leg and the drive assembly. The drive assembly permits instantaneous movement of the first leg and the second leg. A first finger is rotatably coupled to the first leg and a second finger is coupled to the second leg. The first finger has a first finger outer surface and the second finger has a second finger outer surface that are generally coplanar with the body outer surface when the first finger and the second finger are in a retracted position.

Abstract: User interface (UI) tiles of conference participants are arranged in a UI of conferencing software according to relative locations of those conference participants within a conference room. Positional information and video data are obtained from one or more video capture devices located within a conference room. Relative locations of conference participants within the conference room are determined based on the positional information, such as by defining coordinates for the conference participants within a coordinate system based on the positional information and the video data and determining the relative locations based on the coordinates. Output configured to cause a client application to arrange UI tiles associated with the conference participants is generated according to the relative locations. The output is then transmitted to one or more client devices to cause a UI of conferencing software at each of those client devices to display the UI tiles in the specified arrangement.

Abstract: Implementations set forth herein allow a user to access a first application in a foreground of a graphical interface, and simultaneously employ an automated assistant to respond to notifications arising from a second application. The user can provide an input, such as a spoken utterance, while viewing the first application in the foreground in order to respond to notifications from the second application without performing certain intervening steps that can arise under certain circumstances. Such intervening steps can include providing a user confirmation, which can be bypassed, and/or time-limited according to a timer, which can be displayed in response to the user providing a responsive input directed at the notification. A period for the timer can be set according to one or more characteristics that are associated with the notification, the user, and/or any other information that can be associated with the user receiving the notification.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a machine learning model. One of the methods includes obtaining a training data set for training a machine learning model, the training data set comprising a plurality of training inputs; determining a plurality of data augmentation policies, wherein each data augmentation policy defines a procedure for processing a training input to generate a transformed training input; for each data augmentation policy, training the machine learning model using the data augmentation policy; determining, for each data augmentation policy, a quality measure of the machine learning model that has been trained using the data augmentation policy; and selecting a final data augmentation policy based using the quality measures of the machine learning models.

Abstract: A memory device with memory cell pairs each having a single continuous channel region, first and second floating gates over first and second portions of the channel region, an erase gate over a third portion of the channel region between the first and second channel region portions, and first and second control gates over the first and second floating gates. For each of the pairs of memory cells, the first region is electrically connected to the second region of an adjacent pair of memory cells in the same active region, and the second region is electrically connected to the first region of an adjacent pair of the memory cells in the same active region.

Abstract: User interface (UI) tiles of conference participants are arranged in a UI of conferencing software according to relative locations of those conference participants within a conference room. Positional information and video data are obtained from one or more video capture devices located within a conference room. Relative locations of conference participants within the conference room are determined based on the positional information, such as by defining coordinates for the conference participants within a coordinate system based on the positional information and the video data and determining the relative locations based on the coordinates. Output configured to cause a client application to arrange UI tiles associated with the conference participants is generated according to the relative locations. The output is then transmitted to one or more client devices to cause a UI of conferencing software at each of those client devices to display the UI tiles in the specified arrangement.

Abstract: An image processing apparatus according to an embodiment includes a hardware processor. The processor receives input of an image and receives input of a query including one or more words. The processor calculates a word heat map for each of the words. The word heat map indicates a degree of relation between the word and each of subregions included in the image. The processor calculates weight of each of the words. The processor calculates, on the basis of the word heat maps and the weights, a query heat map indicating a degree of relation between the query and each of the subregions included in the image. The processor detects an image region from the image on the basis of the query heat map and outputs the image region. The image region corresponds to an object related to the query.

Abstract: Example embodiments of aspects of the present disclosure provide an example computer-implemented method for improved prompting of a machine-learned model. The example method can include obtaining an instructive sequence descriptive of an instructive query, an instructive response, and an instructive trace of intermediate states from the instructive query to the instructive response. The example method can include inputting, to a machine-learned model, the instructive sequence and an operative query, wherein the machine-learned model is configured to process the operative query with attention over the instructive sequence. The example method can include generating, using the machine-learned model and responsive to the operative query, an operative response.

Abstract: A system for determining risky events includes an input interface and a processor. The input interface is for receiving sensor data on environmental conditions. The processor is for determining whether the environmental conditions indicate an increase in event probability and, in the event that environmental conditions indicate the increase in the event probability, adjusting an event detection threshold.

Abstract: Methods, apparatus and systems for wireless sensing are described. In one example, a described system in a wireless data communication network comprises: a transmitter configured to transmit a time series of at least one wireless sounding signal (WSS) based on a standard wireless network protocol associated with the wireless data communication network, and a receiver. The wireless data communication network comprises a physical (PHY) layer, a medium access control (MAC) layer, and at least one higher layer. The receiver is configured to: receive the time series of at least one WSS (TSWSS) based on the standard wireless network protocol through a wireless channel of a venue, and perform a plurality of wireless sensing measurements based on the received TSWSS to obtain sensing measurement results based on the standard wireless network protocol.

Abstract: Methods of depositing material onto substrate comprising: depositing a first seed material onto a wafer substrate, the wafer substrate having a face that defines a normal to the substrate, wherein the first seed material is deposited at a pressure of 10 to 20 mTorr to form a pre-seed layer on the wafer substrate, wherein the pre-seed layer has a surface roughness from 1 to 10 nm; depositing a second seed material onto at least a portion of the pre-seed layer at an off-normal incidence angle to form a seed layer on at least a portion of the pre-seed layer; and depositing a bulk piezoelectric material onto at least a portion of the seed layer to form a bulk piezoelectric layer having a c-axis tilt of 35 degrees or greater and a surface roughness of 4.5 nm or less. Structures and bulk acoustic wave resonators containing same are also included.

Abstract: A method and system for auto-curating a media are provided. Media content is received over the network interface. A set of markers is identified for the media content, each marker corresponding to one of a plurality of visible and audible cues in the media content. Segments in the media content are identified based on the identified set of markers. An excitement score is computed for each segment based on the identified markers that fall within the segment. A highlight clip is generated by identifying segments having excitement scores greater than a threshold.

Abstract: A dynamic asset spawning engine is provided in a data processing system for dynamically spawning assets in a video game based on one or more news feeds. The dynamic asset spawning engine semantically tags a library of video game assets with game tag descriptors. The dynamic asset spawning engine parses and processes the one or more news feeds to identify supplemental tag descriptors and spawning parameters for one or more video game assets in the library of video game assets. The dynamic spawning engine assigns the identified supplemental tag descriptors and spawning parameters to the one or more video game assets. The video game spawns the one or more game assets in the video game in accordance with the supplemental tag descriptors and spawning parameters.

Abstract: A computer-implemented method for auto-splitting and classifying an input document into one or more sub-documents using a machine learning system is described. The machine learning system includes a visual splitting neural network, an optical character recognition sub-system, a title classifier, a document classifier, and a grouper sub-system. The method includes: receiving a visual input representing a plurality of pages of the input document; classifying, using the visual splitting neural network, each page of the input document into a respective template of a plurality of templates; for each page of the input document, determining a final document type that the page belongs to; and grouping, using a grouper sub-system, the plurality of pages of the input document into one or more sub-documents based on (i) the respective template of each page and (ii) the respective final document type that each page belongs to.