Abstract: A machine learning system can include a symbolic memory network and a secondary memory network. The machine learning system can retrieve a trace from a secondary memory network and reprocess the trace. The machine learning system can generate an analog by reprocessing the trace using at least one of input data obtained from an environment, or at least one analog obtained from the symbolic memory network. The machine learning system can determine that the generated analog satisfies a storage condition for the symbolic memory network. The machine learning system can provide instructions to take an action in the environment based on a subsequent retrieval of the generated analog from the symbolic memory network.

Abstract: A machine learning system can include a symbolic representation generator and an inference component. The symbolic representation can generate a symbolic memory network using input obtained from an environment. The inference component can generate a causal model by experimenting upon components, operations, and parameters of the symbolic representation generator. The causal model can influence the generation of the symbolic memory network. In particular, the machine learning system can identify an analog in the symbolic memory network as being entangled, identify a disentangled version of the identified entangled analog in the symbolic memory network, and, in response to the identification of the disentangled version of the identified entangled analog, release the identified entangled analog from the symbolic memory network and provide instructions to take an action in an environment based on the disentangled version of the identified entangled analog.

Abstract: A symbolic machine learning system can include a symbolic representation generator, an inference component, and an interface component. The symbolic representation generator can repeatedly receive inputs associated with observed objects and computes a symbolic memory network representative of associations involving the observed objects. The inference component can repeatedly perturb components, operations, or parameters of the symbolic representation generator and computes a causal model of effects of the perturbations. The interface component can output signals generated based on the causal model and concerning the associations involving the observed objects.

Abstract: A machine learning system including a symbolic representation generator can generate a first relationship analog through non-interactive learning. The symbolic representation generator can generate a second relationship analog through interactive experimentation using the first relationship analog. The interactive experimentation can include accumulating analogs during an experimental unit and specifying a goal. The interactive experimentation can include retrieving the first relationship analog based on the specified goal and the accumulated analogs. The interactive experimentation can include generating the second relationship analog using at least one of relational learning or comparison-based refinement, where the generation can use the first relationship analog and the accumulated analogs.

Abstract: A machine learning system can determine symbolic representation generator parameters. The machine learning system can select and configure an experimental design. The machine learning system can assign symbolic representation experimental conditions (SRECs) to a policy according to the experimental design. During an experimental unit, the machine learning system can match SRECs assigned to the policy to analogs in a symbolic representation generated by the symbolic representation generator using input obtained from an environment according to the symbolic representation generator parameters. The machine learning system can provide instructions to take actions in the environment based on the matched SRECs and to update the symbolic representation generator parameters based on a result of the experimental unit.

Abstract: A method includes detecting, via object detection hardware, a portion of at least one of a plurality of objects, receiving object attributes for the at least one of a plurality of objects, providing at least one contact area, based upon the object attributes, on each of a plurality of object representations corresponding to each of the at least one of a plurality of objects, providing a surface representation, displaying, via display hardware, the plurality of object representations each residing upon the surface representation, and displaying at least one support area on the surface representation corresponding to the at least one contact area associated with the plurality of object representations.

Abstract: The disclosure describes systems of navigating a hazardous environment. The system includes personal protective equipment (PPE) and computing device(s) configured to process sensor data from the PPE, generate pose data of an agent based on the processed sensor data, and track the pose data as the agent moves through the hazardous environment. The PPE may include an inertial measurement device to generate inertial data and a radar device to generate radar data for detecting a presence or arrangement of objects in a visually obscured environment. The PPE may include a thermal image capture device to generate thermal image data for detecting and classifying thermal features of the hazardous environment. The PPE may include one or more sensors to detect a fiducial marker in a visually obscured environment for identifying features in the visually obscured environment. In these ways, the systems may more safely navigate the agent through the hazardous environment.

Abstract: A method includes detecting, via object detection hardware, a portion of at least one of a plurality of objects, receiving object attributes for the at least one of a plurality of objects, providing at least one contact area, based upon the object attributes, on each of a plurality of object representations corresponding to each of the at least one of a plurality of objects, providing a surface representation, displaying, via display hardware, the plurality of object representations each residing upon the surface representation, and displaying at least one support area on the surface representation corresponding to the at least one contact area associated with the plurality of object representations.

Abstract: Augmented reality (AR) devices, systems and methods are provided to argument captured images of objects of interest. Image data can be obtained for an object of interest by an imaging device of an ear-worn device worn by a user. Augmenting information is generated to augment onto an image of the object. The augmented image is adjusted based on the detection on whether the object is in a field of view (FOV) of the imaging device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimizing a process of manufacturing a product. In one aspect, the method comprises repeatedly performing the following: i) selecting a configuration of input settings for manufacturing a product, based on a causal model that measures causal relationships between input settings and a measure of a quality of the product; ii) determining the measure of the quality of the product manufactured using the configuration of input settings; and iii) adjusting, based on the measure of the quality of the product manufactured using the configuration of input settings, the causal model.

Abstract: The present disclosure includes in one instance an optical article comprising a data rich plurality of retroreflective elements that are configured in a spatially defined arrangement, where the plurality of retroreflective elements comprise retroreflective elements having at least two different retroreflective properties, and where data rich means information that is readily machine interpretable. The present disclosure also includes a system comprising the previously mentioned optical article, an optical system, and an inference engine for interpreting and classifying the plurality of retroreflective elements wherein the optical system feeds data to the inference engine.

Abstract: Optical articles including a spatially defined arrangement of a plurality of data rich retroreflective elements, wherein the plurality of retroreflective elements comprise retroreflective elements having at least two different retroreflective properties and at least two different optical contrasts with respect to a background substrate when observed within an ultraviolet spectrum, a visible spectrum, a near-infrared spectrum, or a combination thereof.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimizing a process of manufacturing a product. In one aspect, the method comprises repeatedly performing the following: i) selecting a configuration of input settings for manufacturing a product, based on a causal model that measures causal relationships between input settings and a measure of a quality of the product; ii) determining the measure of the quality of the product manufactured using the configuration of input settings; and iii) adjusting, based on the measure of the quality of the product manufactured using the configuration of input settings, the causal model.

Abstract: Systems and methods of path-based mapping and routing are provided. Translation information and absolute information of mobile objects in environments are determined based on a fusion of sensing data from a radar and an inertial measurement unit (IMU) including a gyroscope and an accelerometer, from which path-based maps and optimal routes can be generated.

Abstract: The present disclosure includes in one instance an optical article comprising a data rich plurality of retroreflective elements that are configured in a spatially defined arrangement, where the plurality of retroreflective elements comprise retroreflective elements having at least two different retroreflective properties, and where data rich means information that is readily machine interpretable. The present disclosure also includes a system comprising the previously mentioned optical article, an optical system, and an inference engine for interpreting and classifying the plurality of retroreflective elements wherein the optical system feeds data to the inference engine.

Abstract: Augmented reality (AR) devices, systems and methods are provided to argument captured images of objects of interest. Image data can be obtained for an object of interest by an imaging device of an ear-worn device worn by a user. Augmenting information is generated to augment onto an image of the object. The augmented image is adjusted based on the detection on whether the object is in a field of view (FOV) of the imaging device.

Abstract: The disclosure describes systems (2) of navigating a hazardous environment (8). The system includes personal protective equipment (PPE) (13) and computing device(s) (32) configured to process sensor data from the PPE (13), generate pose data of an agent (10) based on the processed sensor data, and track the pose data as the agent (10) moves through the hazardous environment (8). The PPE (13) may include an inertial measurement device to generate inertial data and a radar device to generate radar data for detecting a presence or arrangement of objects in a visually obscured environment (8). The PPE (13) may include a thermal image capture device to generate thermal image data for detecting and classifying thermal features of the hazardous environment (8). The PPE (13) may include one or more sensors to detect a fiducial marker (21) in a visually obscured environment (8) for identifying features in the visually obscured environment (8).

Abstract: The disclosure describes systems of navigating a hazardous environment. The system includes personal protective equipment (PPE) and computing device(s) configured to process sensor data from the PPE, generate pose data of an agent based on the processed sensor data, and track the pose data as the agent moves through the hazardous environment. The PPE may include an inertial measurement device to generate inertial data and a radar device to generate radar data for detecting a presence or arrangement of objects in a visually obscured environment. The PPE may include a thermal image capture device to generate thermal image data for detecting and classifying thermal features of the hazardous environment. The PPE may include one or more sensors to detect a fiducial marker in a visually obscured environment for identifying features in the visually obscured environment. In these ways, the systems may more safely navigate the agent through the hazardous environment.

Abstract: In general, techniques are described for a personal protective equipment (PPE) management system (PPEMS) that uses images of optical patterns embodied on articles of personal protective equipment (PPEs) to identify safety conditions that correspond to usage of the PPEs. In one example, an article of personal protective equipment (PPE) includes a first optical pattern embodied on a surface of the article of PPE; a second optical pattern embodied on the surface of the article of PPE, wherein a spatial relation between the first optical pattern and the second optical pattern is indicative of an operational status of the article of PPE.

Abstract: Example systems are described in which a personal protective equipment (PPE) respirator device includes one or more sensor arranged within the PPE to detect exhalation breath temperature of a user, and a computing device configured to generate, based on the detected exhalation breath temperature, a metric indicative of core body temperature of the user wearing the PPE.