Abstract: A facial-interface system for a head-mounted display may include a facial interface and a facial-interface adjustment apparatus. The facial-interface adjustment apparatus may include (1) a coupling mount for coupling the facial-interface adjustment apparatus to a head-mounted-display housing, (2) an adjustment protrusion that is movable between a plurality of holding positions relative to the coupling mount, the adjustment protrusion having a facial-interface coupling portion coupled to the facial interface, and (3) a spring that biases the adjustment protrusion in a biasing direction toward the facial interface such that the adjustment protrusion moves between separate holding positions of the plurality of holding positions following application of a compressing force to the adjustment protrusion compressing the spring in a compressing direction opposite the biasing direction. Various other systems, devices, and methods are also disclosed.

Abstract: A light-emitting diode includes a first semiconductor region of one of p- or n-conductivity types, a second semiconductor region of the other one of p- or n-conductivity types, forming a p-n junction with the first semiconductor region, and a quantum well layer at the p-n junction between the first and second semiconductor regions. A hyperbolic metamaterial structure is provided in the second semiconductor region. The hyperbolic metamaterial structure is coupled to the quantum well layer for extracting light from the quantum well layer. The hyperbolic metamaterial structure may be patterned to provide an array of nanoantennas to apodize the emitted beam, and to control the polarization state of the emitted beam.

Abstract: A system predicts user intent to take an action and delivers content items to the user that match that intent. A plurality of features or attributes for each tracking pixel in a set of tracking pixels can be acquired based on content items and landing pages associated with each tracking pixel. For example, features for a tracking pixel can be determined based on information associated with a content item that enabled a user to access a landing page from which the tracking pixel was fired or triggered. In this example, features for the tracking pixel can also be determined based on information associated with the landing page. The features for the tracking pixels can be utilized to train a machine learning model. The machine learning model can be trained to predict whether or not a particular user intends to produce a conversion (e.g., make a purchase).

Abstract: Ion implantation is used to fabricate an optical device having a varying refractive index. The optical device can include a substrate with a material disposed on the substrate. A refractive index of the material is changed by ion implantation. The material can also be etched or imprinted. The optical device can be used in a virtual-reality system or augmented-reality system to provide angular selectivity from a display to a user's eye.

Abstract: Disclosed herein is a method for using a neural network across multiple devices. The method can include receiving, by a first device configured with a first one or more layers of a neural network, input data for processing via the neural network implemented across the first device and a second device. The method can include outputting, by the first one or more layers of the neural network implemented on the first device, a data set that is reduced in size relative to the input data while identifying one or more features of the input data for processing by a second one or more layers of the neural network. The method can include communicating, by the first device, the data set to the second device for processing via the second one or more layers of the neural network implemented on the second device.

Abstract: Disclosed herein includes a system, a method, and a device for asymmetrical scaling factor support for negative and positive values. A device can include a circuit having a shift circuitry and multiply circuitry. The circuit can be configured to perform computation for a neural network, including multiplying, via the multiply circuitry, a first value and a second value. The circuit can be configured to perform computation for a neural network, including shifting, via the shift circuitry, a result of the multiplying by a determined number of bits. The circuit can be configured to perform computation for a neural network, including outputting the result of the multiplying when a sign bit of the first value is negative, and a result of the shifting when the sign bit of the first value is positive.

Abstract: Disclosed herein includes a system, a method, and a device for multiply-accumulate operation. In one aspect, an input operand is received by control circuitry. In one aspect, the control circuitry determines a sparsity of the input operand, where the sparsity may indicate whether a value of the input operand has a predetermined value or not. In one aspect, the control circuitry determines a stationarity of the input operand, where the stationarity may indicate whether the value of the input operand changes over one or more clock cycles. In one aspect, the input operand is provided to multiply-accumulate circuitry as an input, according to the determined sparsity and stationarity of the input operand.

Abstract: Disclosed herein includes a system, a method, and a device for early-exit from convolution. In some embodiments, at least one processing element (PE) circuit is configured to perform, for a node of a neural network corresponding to a dot-product operation with a set of operands, computation using a subset of the set of operands to generate a dot-product value of the subset of the set of operands. The at least one PE circuit can compare the dot-product value of the subset of the set of operands, to a threshold value. The at least one PE circuit can determine whether to activate the node of the neural network, based at least on a result of the comparing.

Abstract: Disclosed herein includes a system, a method, and a device for pipelined parallelism to accelerate distributed learning network graph. First data for a first layer of a neural network may be stored in memory. First circuitry including a first plurality of processing element (PE) circuits may read the first data from the memory and perform computation for the first layer of the neural network using the first data to generate second data. The first circuitry includes a plurality of buffers for outputting the generated second data as input to second circuitry to perform computation for a second layer of the neural network. The second circuitry includes a second plurality of PE circuits configured to perform computation for the second layer of the neural network using the second data.

Abstract: Disclosed herein includes a system, a method, and a device for reading and writing sparse data in a neural network accelerator. A plurality of slices can be established to access a memory having an access size of a data word. A first slice can be configured to access a first side of the data word in memory. Circuitry can access a mask identifying byte positions within the data word having non-zero values. The circuitry can modify the data word to have non-zero byte values stored starting at an end of the first side, and any zero byte values stored in a remainder of the data word. A determination can be made whether a number of non-zero byte values is less than or equal to a first access size of the first slice. The circuitry can write the modified data word to the memory via at least the first slice.

Abstract: An online system receives a request for a video to be presented by the online system to a target user. The online system determines whether to insert secondary content into the video. For such a determination, the online system identifies a position in the video for inserting secondary content. Further, the online system determines a loss score and a gain score. The loss score measures a loss of interaction by the target user if the secondary content were inserted. The gain score includes a monetary compensation to be received by the online system for inserting the secondary content at the identified position. The online system compares the loss score and the gain score. Based on the gain score offsetting the loss score, the online systems modifies the video by inserting the secondary content at the identified position and provides the modified video for display to the target user.

Abstract: An online system presents a content item to users and receives selections of reaction icons from the users. The online system generates a background animation with the selected reaction icons and a foreground animation to be layered on top of the background animation. The online system sends the background and foreground animations to a client device to be cached. Further, the online system presents the content item to a viewing user associated with the client device and receives a selection of a reaction icon from the viewing user. The online system selects a subset of the users based on the viewing user's affinity to the users, retrieves images of the selected users, and send the images to the client device. The client device customizes the background and foreground animations based on the images and the viewing user's reaction icon to generate a compound animation for display to the viewing user.

Abstract: A speaker assembly presents audio content to an ear canal of a user. A speaker of the speaker assembly generates positive and negative acoustic pressure waves. A first vent assembly of the speaker assembly ports the positive acoustic pressure waves to an entrance of the ear canal of the user, whereas a second vent assembly ports the negative acoustic pressure waves to an area behind a pinna of the user. The first and second vent assembly are configured to provide improved audio content playback to the user.

Abstract: An optical device is provided. The optical device includes an optical grating. The optical grating includes two electrodes arranged opposite to each other and a liquid crystal (LC) composition sandwiched between the electrodes. The two electrodes provide a driving voltage to the optical grating. The LC mixture includes a host LC and one or more LC dimers dissolved as a guest in the host LC. The host LC and the one or more LC dimers have respective dielectric anisotropies of opposite signs in nematic phase. A net dielectric anisotropy of the LC mixture is substantially neutral at a predetermined temperature.

Abstract: A social networking system obtains information describing one or more interactions performed by users of the social networking system with various applications. The information identifies users, applications, and interactions by users with applications. Based on the obtained information, the social networking system determines likelihoods of a selected user performing another type of interaction with each application in a set of applications. The set of applications are ranked based on the determined likelihoods, and one or more of the applications from the set are selected based on the ranking. Information describing the selected one or more applications is then provided to the selected user.

Abstract: An augmented reality (AR) headset includes a depth camera assembly that combines stereo imaging with structured light (SL) to generate depth information for an area of interest. The depth camera assembly includes at least two image capture devices and a SL illuminator and determines an imaging mode based on a signal to noise ratio or spatial variance of images captured by one or more of the cameras. Different imaging modes correspond to different operation of one or more image capture devices and the SL illuminator. The depth camera assembly includes different ranges of signal to noise ratios that each correspond to an imaging mode, and the depth camera assembly configures the image capture devices and the SL illuminator based on an imaging mode associated with a range of signal to noise ratios including the signal to noise ratio of a captured image.

Abstract: An artificial reality system is described that renders, presents, and controls user interface elements within an artificial reality environment, and performs actions in response to one or more detected gestures of the user. The artificial reality system can include a menu that can be activated and interacted with using one hand. In response to detecting a menu activation gesture performed using one hand, the artificial reality system can cause a menu to be rendered. A menu sliding gesture (e.g., horizontal motion) of the hand can be used to cause a slidably engageable user interface (UI) element to move along a horizontal dimension of the menu while horizontal positioning of the UI menu is held constant. Motion of the hand orthogonal to the menu sliding gesture (e.g., non-horizontal motion) can cause the menu to be repositioned. The implementation of the artificial reality system does require use of both hands or use of other input devices in order to interact with the artificial reality system.

Abstract: A near-eye display assembly includes an electronic display for providing image light, a circular polarizer disposed downstream of the electronic display for circularly polarizing the image light, and a pancake lens disposed downstream of the circular polarizer for conveying the image light to an eyebox of the near-eye display. The circular polarizer includes a quarter-wave waveplate disposed between the electronic display and the pancake lens, and spaced apart from the electronic display and the pancake lens, to additionally defocus ghost image artifacts at the eyebox.

Abstract: An optical system includes a grating including at least one substrate and a grating structure coupled to the at least one substrate. The grating structure is configured to diffract a first light having an incidence angle within a predetermined range. The optical system also includes a polarizer configured to transmit the first light diffracted by the grating structure and block a second light reflected by a surface of the at least one substrate.

Abstract: A robotic platform may include a chassis, left and right wheel assemblies, and a controller. The left and right wheel assemblies may include a caster wheel, a motor, a shaft, and a bevel gear. The wheel may be mounted to an axle for rotation about a drive axis and steering about a steering axis. The drive shaft may have one end coupled to the axle and another end wrapped by a respective belt to control rotation of the shaft about the steering axis. The bevel gear may couple the shaft to the axle so rotation of the shaft about the steering axis controls rotation of the wheel about the drive axis to drive the platform in a substantially horizontal direction. The controller may control the left and right drive motors independently, to provide differential drive. Various other assemblies, robots, and methods are also disclosed.