Abstract: A system for servicing watercraft includes one or more waterborne platforms. Each waterborne platform includes an electric power supply, a driving system for moving the waterborne platform in a body of water, a watercraft interfacing system configured to at least supply electric power to an electrically-powered watercraft, and a control interface configured to exchange data with a controller. The controller is configured to: receive input data, determine respective destination locations for the waterborne platforms to supply electric power to the electrically-powered watercraft, and send control data that includes data indicating the destination locations to the waterborne platforms.

Abstract: The present disclosure provides systems, methods, and computer program products for generating a digital representation of a system from engineering documents of the system comprising one or more schematics and a components table. An example method can comprise (a) classifying, using a deep learning algorithm, (i) each of a plurality of symbols in the one or more schematics as a component and (ii) each group of related symbols as an assembly, (b) determining connections between the components and the assemblies, (c) associating a subset of the components and the assemblies with entries in the components table; and (d) generating the digital representation of the system from the components, the assemblies, the connections, and the associations. The digital representation of the system can comprise at least a digital model of the system and a machine-readable bill of materials.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for automated camera positioning for feeding behavior monitoring. In some implementations, a system obtains an image of a scene, a spatial model that corresponds to a subfeeder, and calibration parameters of a camera, the system determines a size of the subfeeder in the image of the scene, the system selects an updated position of the camera relative to the subfeeder, the system provides the updated position of the camera relative to the subfeeder to a winch controller, and the system moves the camera to the updated position.

Abstract: Low-noise electric lawnmowers comprise curved cutting blades. In some examples, a lawnmower comprises a driving unit and a mowing unit with a mowing-unit deck. The mowing-unit deck comprises a set of blade units (e.g., 3 units) and a discharge conduit. Each blade unit comprises a blade enclosure and a cutting blade with a cutting edge having a curved shape. For example, the cutting edge, at a point furthest away from the blade center axis, may form an approach angle of greater than 90° but be less than 90° closer to the blade center. This curving aspect allows operating the blade at lower rotating speeds (e.g., less than 3,000 RPM), which reduces the turbulence and noise. In some examples, the blade enclosure forms an enclosure tunnel such that the tunnel's cross-sectional area increases from the tunnel's inlet to the outlet.

Abstract: One embodiment of the present invention sets forth a technique for evaluating labeled data. The technique includes selecting, from a set of labels for a data sample, a subset of the labels representing non-outliers in a distribution of values in the set of labels. The technique also includes aggregating the subset of the labels into a benchmark for the data sample. The technique further includes generating, based on a comparison between the benchmark and an additional label, a benchmark score associated with the data sample, and generating a set of performance metrics for labeling the data sample based on the benchmark score.

Abstract: An apparatus for generating a lip sync image according to a disclosed embodiment has one or more processors and a memory which stores one or more programs executed by the one or more processors. The apparatus includes a first artificial neural network model configured to generate an utterance synthesis image by using a person background image and an utterance audio signal corresponding to the person background image as an input, and generate a silence synthesis image by using only the person background image as an input, and a second artificial neural network model configured to output, from a preset utterance maintenance image and the first artificial neural network model, classification values for the preset utterance maintenance image and the silence synthesis image by using the silence synthesis image as an input.

Abstract: The disclosed embodiments relate to a system that generates and executes a deep neural network (DNN) based on target runtime parameters. During operation, the system receives a trained original model and a set of target runtime parameters for the DNN, wherein the target runtime parameters are associated with one or more of the following for the DNN: desired operating conditions, desired resource utilization, and desired accuracy of results. Next, the system generates a context-specific model based on the original model and the set of target runtime parameters. The system also generates an operational plan for executing both the original model and the context-specific model to meet requirements of the target runtime parameters. Finally, the system controls execution of the original model and the context-specific model based on the operational plan.

Abstract: This disclosure provides novel methods and systems for indexing embedding vectors representing disjoint classes at above-billion scale for fast high-recall retrieval. The disclosed methods and systems solved above-billion scale image (e.g., face image) search problems by building a new vector database, the new Nearest Neighbor Database (NNDB) based on a novel architecture for indexing vectors.

Abstract: The present disclosures provides for a system and a method for evaluating brain trauma exposure from a motion capture source. Preferably, the system is comprised of a motion capture source engaged with a non-transitory computer readable medium such as a processor. The motion capture source is adapted to view or record images or videos of human movement. The processor determines and extracts key characteristics of a head impact and predicts a brain response based on such characteristics. The system is then able to generate and provide a brain trauma assessment output based on the brain response. A method is also disclosed, which extracts key characteristics from the motion capture source, predicts a brain response based on such key characteristics and provides a brain trauma exposure output based on the brain response.

Abstract: Systems and methods for implementing trained embedding mappings for improved retrieval augmented generation are disclosed. A system can maintain a dataset comprising a first set of embeddings corresponding to a first embeddings space and stored in association with a set of query results for the first set of embeddings. The set of query results can correspond to a second embeddings space. The system can train a transformation data structure using the first set of embeddings and the set of query results. The transformation data structure can be used to transform the first set of embeddings to the second embeddings space. The system can execute a search operation for the second embeddings space by applying the transformation data structure to a second set of embeddings corresponding to the first embeddings space.

Abstract: One embodiment of the present invention sets forth a technique for extracting data from a document. The technique includes determining, via execution of one or more machine learning models, a first set of bounding boxes for a first set of values associated with a first key within the document and a second set of bounding boxes for a second set of values associated with a second key within the document. The technique also includes generating a first set of mappings between a list of items in the document and the first set of bounding boxes and a second set of mappings between the first and second sets of bounding boxes based on locations of the bounding boxes. The technique further includes determining, for a given item, one or more associated bounding boxes in the first and second sets of bounding boxes based on the mappings.

Abstract: Generating a subsetting traversal plan includes receiving a set of source tables. It further includes sorting the set of source tables based at least in part on dependency relationships among tables in the set of source tables. It further includes generating a traversal order based at least in part on the sorting of the set of source tables. It further includes providing the traversal order as output.

Abstract: To effectively operate in an increasingly prolific Internet of Things environment, one best has the ability to message, whether device status, actions taken, recommendations for actions or other information to those with a need to know or respond to device-execution instructions. Traditional messaging such as texting and e-mail typically are limited in that they can neither provide automatic audio output to the recipient nor can they may be automatically relayed or contain execution instructions for target devices. The present invention is an appliance that communicates with other IoT devices that provides Extended-Capability Messaging with these capabilities and can include AI and optionally configured in a secure version. Artificial-Intelligence elements of a variety of types can be included in Extended-Capability Messaging Appliances and these and other AI elements constitute a Hybrid AI Backbone.

Abstract: The present disclosure provides system, methods, and computer program products for predicting and detecting anomalies in a subsystem of a system. An example method may comprise (a) determining a first plurality of tags that are indicative of an operational performance of the subsystem. The tags can be obtained from (i) a plurality of sensors in the subsystem and (ii) a plurality of sensors in the system that are not in the subsystem. The method may further comprise (b) processing measured values of the first plurality of tags using an autoencoder trained on historical values of the first plurality of tags to generate estimated values of the first plurality of tags; (c) determining whether a difference between the measured values and estimated values meets a threshold; and (d) transmitting an alert that indicates that the subsystem is predicted to experience an anomaly if the difference meets the threshold.

Abstract: A method for binary distance transform includes identifying a first binary vector and a second binary vector, the first binary vector and the second binary vector both having a binary dimension m, the binary dimension m indicating a number of bits. The method also includes determining a first distance between the first binary vector and the second binary vector, the first distance including a count of differing bits between the first binary vector and the second binary vector. The method also includes transforming the first distance to a second distance. The method also includes executing, by one or more processors, a machine-readable instruction in view of the second distance.

Abstract: A method includes identifying a feature vector with N features to transform into M bits. The feature vector including one or more values, where each value corresponds to a characteristic of an object. The method includes applying a transform to the feature vector to create a transformed feature vector. The method further includes quantizing the transformed feature vector to generate a binarized feature vector of M bits.

Abstract: An ethical AI development platform operates by: facilitating the development of a training dataset; providing access to auto machine learning tools to facilitate the development of an AI model; providing access to machine learning analysis tools to facilitate the evaluation of the AI model, wherein the machine learning analysis tools include an AI moral insight model that generates predicted moral score data associated with the AI model, wherein the AI moral insight model is trained based on at least one-hundred thousand sets of training data, wherein the training data includes reviewed media content and moral score data associated with moral indices, wherein the moral indices include: a degree of immorality/morality, a degree of betrayal/loyalty, a degree of degradation/sanctity, a degree of harm/care a degree of cheating/fairness and a degree of subversion/authority; and providing access to a version control repository for storing versions of the training dataset and the AI model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for a lighting controller for sea lice detection. In some implementations, fish are contained within an elliptical tank filled with water. An imaging station located on the elliptical tank is used to capture an image of a fish from which image analysis can be performed to detect sea lice or other skin features, including lesions, on the fish. Pairs of imaging assemblies coordinate pulsing light of at least a first and a second color and capturing images of the fish while the fish is illuminated by at least the first and the second color. By using the captured images with different color light, computers can detect features on the body of a fish including sea lice, skin lesions, shortened operculum or other physical deformities and skin features. Detection results can aid in mitigation techniques or be stored for analytics.

Abstract: A system for designing a robot for performing a specified function includes a library, a robot design server, and a robot data server. The library stores data about modules and accessories for a robot and has at least unique IDs, dimensions, and firmware for operation of each module and accessory. The robot design server receives a set of functions and requirements of the new robot, finds a set of modules or accessories which match at least one of the set of functions and requirements, performs a set of validation tests against a set of criteria, identifies at least one design which passed the set of validation tests, and generates a list of modules and accessories and their unique IDs of a selected design and a connection map for the selected design. The robot data server downloads to the assembled robot the data and metadata of its modules and accessories.

Abstract: Apparatus and associated methods relate to emergency stroke detection and classification. In an illustrative example, a stroke detection device may include an ensemble stroke classification model (ESCM). The ESCM may, for example, include class-specific model sets applicable for at least four classes of features, and a general model set applicable for all classes of features. Each model set, for example, may be stacked with multiple class-specific models for each of a corresponding group of architectures. The stroke detection device may, for example, extract predetermined features from a rolling window of a first predetermined duration of EEG data. The predetermined features are extracted and combined into a 1-D input vector. By applying the input vector, the stroke detection device may generate a binary stroke prediction result. Various embodiments may advantageously accurately predict whether a patient is experiencing a stroke within a finite time to assist an emergency service personnel.