Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a student neural network. In one aspect, there is provided a method comprising: processing a training input using the student neural network to generate an output for the training input; processing the student neural network output using a discriminative neural network to generate a discriminative score for the student neural network output, wherein the discriminative score characterizes a prediction for whether the network input was generated using: (i) the student neural network, or (ii) a brain emulation neural network; and adjusting current values of the student neural network parameters using gradients of an objective function that depends on the discriminative score for the student neural network output.

Abstract: Methods, systems, and computer-readable media that implement autonomous seagoing power replenishment watercraft. An example system includes a plurality of marine vessels; a plurality of watercraft, each watercraft of the plurality of watercraft including a rechargeable electrical power supply and being configured to operate in: a first mode in which the watercraft awaits an assignment to provide electrical energy to a marine vessel of the plurality of marine vessels; a second mode in which the watercraft performs operations including keeping station with an assigned marine vessel and providing electrical energy to the assigned marine vessel from the power supply; and a third mode in which the watercraft recharges the power supply from a charging station. The system includes a controller configured to perform operations comprising: transmitting, to a first watercraft, an instruction indicating an assignment of the first watercraft to provide electrical energy to a first marine vessel.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for predicting object pose. In one aspect, a method includes receiving an image of an object having one or more feature points; providing the image as an input to a neural network subsystem trained to receive images of objects and to generate an output including a heat map for each feature point; applying a differentiable transformation on each heat map to generate respective one or more feature coordinates for each feature point; providing the feature coordinates for each feature point as input to an object pose solver configured to compute a predicted object pose for the object, wherein the predicted object pose for the object specifies a position and an orientation of an object; and receiving, at the output of the object pose solver, a predicted object pose for the object in the image.

Abstract: A method includes receiving sensor data from a plurality of robot sensors on a robot. The method includes generating a depth map that includes a plurality of pixel depths. The method includes determining, for each respective pixel depth, based on the at least one robot sensor associated with the respective pixel depth, a pixelwise confidence level indicative of a likelihood that the respective pixel depth accurately represents a distance between the robot and a feature of the environment. The method includes generating a pixelwise filterable depth map for a control system of the robot. The pixelwise filterable depth map is filterable to produce a robot operation specific depth map. The robot operation specific depth map is determined based on a comparison of each respective pixelwise confidence level with a confidence threshold corresponding to at least one operation of the robot controlled by the control system of the robot.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting a neural network architecture for performing a machine learning task. In one aspect, a method comprises: obtaining data defining a synaptic connectivity graph representing synaptic connectivity between neurons in a brain of a biological organism; generating data defining a plurality of candidate graphs based on the synaptic connectivity graph; determining, for each candidate graph, a performance measure on a machine learning task of a neural network having a neural network architecture that is specified by the candidate graph; and selecting a final neural network architecture for performing the machine learning task based on the performance measures.

Abstract: A photonic integrated circuit comprises an optical deinterleaver, including an input region, a dispersive region, and at least two output regions. The input region is adapted to receive an input optical signal including a plurality of channels. The dispersive region is optically coupled to the input region to receive the input optical signal. The dispersive region includes an inhomogeneous arrangement of a first material and a second material to structure the dispersive region to separate the input optical signal into a plurality of multi-channel optical signals, including a first multi-channel optical signal and a second multi-channel optical signal. The at least two output regions, include a first out region and a second output region optically coupled to the dispersive region. The first output region is positioned to receive the first multi-channel optical signal and the second output region is positioned to receive the second multi-channel optical signal.

Abstract: A system, including an optical imaging assembly configured to image a sample at an object plane to an image plane; an image sensor arranged at the image plane and configured to capture images of the sample for a field of view of the system; a light source configured to emit light having a wavelength, ?; a spatial light modulator (SLM) arranged to receive the light emitted from the light source and to provide a spatially modulated light pattern; one or more optical elements arranged to receive the spatially modulated light pattern from the SLM and to direct the spatially modulated light pattern to the image plane; and an electronic controller in communication with the image sensor and the spatial light modulator, the electronic controller being programmed to identify one or more targets in the field of view of the optical imaging assembly and to control the spatial light modulator to selectively direct light from the light source to the one or more targets identified by the electronic controller.

Abstract: Methods, systems, apparatus, and computer-readable media for data band selection using machine learning. In some implementations, image data comprising information for each of multiple wavelength bands is obtained. A multi-layer neural network is trained using the image data to perform one or more classification or regression tasks. A proper subset of the wavelength bands is selected based on parameters of a layer of the trained multi-layer neural network, where the parameters were determined through training of the multi-layer neural network using the image data. Output is provided indicating that the selected wavelength bands are selected for the one or more classification or regression tasks.

Abstract: The present disclosure relates to in vitro experiments and in silico computation and machine-learning based techniques to iteratively improve a process for identifying binders that can bind a target. Particularly, aspects of the present disclosure are directed to obtaining initial sequence data, identifying, by a first machine-learning model having model parameters learned from the initial sequence data, a first set of aptamer sequences, obtaining, using an in vitro binding selection process, subsequent sequence data including sequences from the first set of aptamer sequences, identifying, by a second machine-learning model having model parameters learned from the subsequent sequence data, a second set of aptamer sequences, determining, using one or more in vitro assays, analytical data for aptamers synthesized from the second set of aptamer sequences, and identifying a final set of aptamer sequences from the second set of aptamer sequences based on the analytical data associated with each aptamer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for fish weight estimation based on fish tracks identified in images. In some implementations, a method includes obtaining images of fish enclosed in a fish enclosure, identifying fish tracks shown in the images of the fish, determining a quality score for each of the fish tracks, selecting a subset of the fish tracks based on the quality scores, determining a representative weight of the fish in the fish enclosure based on weights of the fish shown in the subset of the fish tracks, and outputting the representative weight for display or storage at a device connected to the one or more processors.

Abstract: A communication device is provided that estimates one or more disturbance values associated with one or more components of the communication device, and adjusts the communication device to change a received power of the output signal. The communication device includes a transmitter having a seed laser configured to provide an amount of bandwidth for an output signal, an Erbium-doped fiber amplifier (EDFA) configured to increase an amplitude of the output signal, and a single mode variable optical attenuator (SMVOA) configured to decrease the amplitude of the output signal.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using a neural network to predict locations of feeders in an electrical power grid. One of the methods includes training a generative adversarial network comprising a generator and a discriminator; and generating, by the generator, from input images, output images with feeder metadata that represents predicted locations of feeder assets, including receiving by the generator a first input image and generating by the generator a corresponding first output image with first feeder data that identifies one or more feeder assets and their respective locations, wherein the one or more feeder assets had not been identified in any input to the generator.

Abstract: A mobile robotic device includes a mobile base and a mast fixed relative to the mobile base. The mast includes a carved-out portion. The mobile robotic device further includes a three-dimensional (3D) lidar sensor mounted in the carved-out portion of the mast and fixed relative to the mast such that a vertical field of view of the 3D lidar sensor is angled downward toward an are in front of the mobile robotic device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium that provides an enhanced synchronization framework. One of the methods includes a primary and a second device that receive configuration information which identifies one or more actions to be performed by the secondary device when it receives specified pulses of a sequence of pulses from the primary device. The primary device transmits a sequence of pulses. The primary and the secondary device receive a particular pulse from the sequence of pulses. The secondary device determines whether the particular pulse satisfies one or more predetermined criteria and generates an instruction based on the determination.

Abstract: A photonic coupler includes an input coupling section, an output coupling section, and a multimode interference (MMI) waveguide section. The input coupling section is adapted to receive an input optical signal along an input waveguide channel. The output coupling section is adapted to output a pair of output optical signals along output waveguide channels. The output optical signals having output optical powers split from the input optical signal. The MMI waveguide section is optically coupled between the input and output coupling sections. Notched waveguide sections may each be disposed between the MMI waveguide section and a corresponding one of the input or output coupling sections and/or the MMI waveguide section may include curvilinear sidewalls.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a synthesized signal. In some implementations, a computer-implemented system obtains generator input data including at least an input signal having one or more first characteristics, processes the generator input data to generate output data including a synthesized signal having one or more second characteristics using a generator neural network, and outputs the synthesized signal to a device. The generator neural network is trained, based on a plurality of training examples, with a discriminator neural network.

Abstract: A latent space is defined to represent sequences using training data and a machine-learning model. The training data identifies sequences of molecules and binding-approximation metrics that characterizes whether the molecules bind to a particular target and/or that approximate an extent to which the molecule is more likely to bind to the particular target than some other molecules. Supplemental training data is accessed that identifies other sequences of other molecules and binding affinity scores quantifying binding strengths between the molecules and the particular target. Projections of representations of the other sequences in the supplemental training data are projected in the latent space using the binding affinity scores. An area or position of interest within the latent space is identified based on the projections. A particular sequence represented within or at the area or position of interest or at the position of interest is identified for downstream processing.

Abstract: A method for matching production of FBA metabolism to supply and demand within a larger production network is described herein. An objective function of FBA metabolism is modified to include an upstream supply generated in upstream sub-units, as well as a downstream demand generated within downstream sub-units in the production network. FBA metabolism and the upstream and downstream sub-units are iteratively solved with updated initial conditions, producing a time series solution to the production network.

Abstract: An alert system is configured to track conditions for a tracking system. The one or more processors of the alert system are configured to receive data in a payload related to characteristics of tracking devices of the tracking system. The one or more processors are then configured to determine a condition timeframe for each condition for a given alert based on the data in the payload. The given alert requires at least a first condition and a second condition of the plurality of conditions. The one or more processors are configured to perform a query for historical data for the first condition based on a timestamp of the second condition that is prior to the first condition, determine an alert timeframe for the given alert based on the condition timeframes and the historical data, and store the given alert and the alert timeframe in a memory of the alert system.

Abstract: The technology employs a state-based power control loop (PCL) architecture to maintain tracking and communication signal-to-noise ratios at suitable levels for optimal tracking performance and data throughput in a free-space optical communication system. Power for a link is adjustable to stay within a functional range of receiving sensors in order to provide continuous service to users. This avoids oversaturation and possible damage to the equipment. The approach can include decreasing or increasing the power to counteract a surge or drop while maintaining a near constant received power at a remote communication device. The system may receive power adjustment feedback from another communication terminal and perform state-based power control according to the received feedback. This can include re-initializing and reacquiring a link with the other communication terminal automatically after loss of power, without human intervention.