Abstract: A method for ground state inference is described. The method includes modeling a material state of a selected material. The method also includes inferring an energy function and a ground state of the selected material according to the modeling of the material state. The method further includes predicting a different material state of the selected material in response to the inferring of the ground state of the material.

Abstract: Systems, methods, and other embodiments described herein relate to downsampling training data so as to simplify the training of models as well as increase the prediction accuracy of the models. In one embodiment, a method includes training a model on a dataset to learn a covariance function, determining a covariance between a selected data value and the dataset using the covariance function, selecting a subset from the dataset based on the covariance, and predicting one or more potential experiments based on the subset.

Abstract: A method for multitask learning based on Hermitian operators is described. The method includes training a multitask machine learning (MTML) model to map an input representation of a material onto a complex wave function state vector. The method also includes inferring, by a trained, MTML model, observable property matrices for each observable property of the material. The method further includes converting the observable property matrices into complex Hermitian operators. The method also includes predicting target properties of the material according to the complex Hermitian operators and the complex wave function state vector.

Abstract: A method for neural network material state prediction is described. The method includes encoding a sequence and interrelationships among events occurring in a simulation and/or experiment in an event-sourced architecture for materials provenance (ESAMP) framework. The method also includes learning an initial state of a material sample in the ESAMP framework. The method further includes sharing a state vector representing the initial state of the material sample with other material samples in the ESAMP framework. The method also includes learning how one or more processes affect the state of the material sample in the ESAMP framework according to the state vector shared with the other material samples in the ESAMP framework.

Abstract: A system for material discovery includes a processor and a memory communicably coupled to the processor. The memory stores an acquisition module, a machine learning module, a duality transform module, and a convex hull module that include instructions that when executed by the processor cause the processor to select a dataset, train a machine learning model to learn a convex function approximating the dataset in a primal space, duality transform hyperplanes of the learned convex function from the primal space to a dual space, learn a convex hull of the duality transformed convex function hyperplanes in the dual space, duality transform at least one hyperplane of the learned convex hull back to the primal space, and predict, based on the at least one duality transformed hyperplane of the learned convex hull, at least one stable material composition within the material space.

Abstract: A machine learning system includes a processor and a memory communicably coupled to the processor. The memory stores machine-readable instructions that, when executed by the processor, cause the processor to select a training dataset comprising training material compositions and tagged material property values, select at least two material property datasets comprising material compositions with corresponding material property values, and embed the training material compositions and the material compositions of the at least two material property datasets into a chemical space of a machine learning module. The memory also stores machine-readable instructions that, when executed by the processor, cause the processor to predict, based at least in part on the training material compositions and the material compositions of the at least two material property datasets embedded in the chemical space, property values for corresponding material compositions in the at least two material property datasets.

Abstract: A system for predicting a one-sided property value for one or more material candidates includes a processor and a memory communicably coupled to the processor. The memory includes a stored acquisition module and a machine learning (ML) module. The acquisition module is configured to select a training data set with a given material property. The training data set includes a first subset of materials having the material property within a predefined range and a second subset of materials having the material property outside the predefined range. Also, the ML module is configured to impute a fixed value for the material property outside the predefined range and train a ML model to predict the material property using imputed fixed value.

Abstract: A machine learning system includes a processor and a memory communicably coupled to the processor. The memory stores an acquisition module, a mapping module, a machine learning module, a fitting module, and a minimization module that include instructions that when executed by the processor cause the processor to: select a training dataset, map the training dataset from an input space to an output space such that the mapped training dataset is convex; train a machine learning model to learn a convex function that approximates the mapped training dataset in the output space; learn a minimum of the convex function; map the minimum of the convex function to the input space; and predict, based at least in part on the minimum of the convex function mapped to the input space, an optimum material property value and a corresponding material composition.

Abstract: System, methods, and other embodiments described herein relate to identifying human-based perception techniques for analyzing a driving scene. In one embodiment, a method includes generating the driving scene as a simulated environment of a vehicle. The method includes modifying the simulated environment according to a visualization algorithm that approximates a machine vision technique to transform the simulated environment into a modified environment with redacted information in comparison to the simulated environment. The method includes displaying the modified environment on an electronic display to an operator to assess how the operator perceives the modified environment when operating the vehicle.

Abstract: Material selection processes can be simulated to determine an optimal material selection process. In iterations: (1) a process selection module can select a material selection module, (2) a machine learning process can be configured to execute the material selection module, (3) the material selection module can: (a) select information about materials having known values of a material property and (b) train the machine learning process to produce the known values in response to a receipt of the information about the materials, and (4) a measure of a performance of the material selection module can be determined with respect to identifying a set of materials that includes the materials for which the known values are in a specific relationship with a threshold criterion for the material property. At a completion of the iterations and based on measures of performances of material selection modules, the optimal material selection process can be determined.

Abstract: Compositions and processes for optimizing oxygen reduction and oxygen evolution reactions are provided. Oxygen reduction and oxygen evolution catalysts include oxide compositions having a general formula a formula A2-xMOy, where x is electrochemically tuned to find optimal A content that delivers the best catalytic performance in a chemical system. The process provides the ability to find the optimal catalytic performance by tuning A and hence, the binding strength of O.

Abstract: Compositions and process for optimizing oxygen reduction and oxygen evolution reactions are provided. Oxygen reduction and oxygen evolution catalysts include oxide compositions having a general formula a formula A2-xMOy, where x is electrochemically tuned to find optimal A content that delivers the best catalytic performance in a chemical system. The process provides the ability to find the optimal catalytic performance by tuning A and hence, the binding strength of O.

Abstract: Embodiments described herein disclose methods and systems for discrete mobile object monitoring. Using location and information from the recognition process, historical action information about a discrete mobile object, and associated object statistical information derived from the historical action information, a hypothesis can be generated for future actions and movements of the discrete mobile object. The information can then be stored for later hypothesis derivation, thus providing a more human-like understanding of the discrete mobile object, useful in a variety of automated tasks.

Abstract: Embodiments described herein disclose methods and systems for dynamic object recognition. Using location and information from the recognition process, historical data for a dynamic object, and associated data from secondary objects, a more complete data set can be generated for the dynamic object. The dynamic object data set can then be stored for later recognition and a more complete and human-like understanding of the dynamic object, useful in a variety of automated tasks.

Abstract: System, methods, and other embodiments described herein relate to selectively processing sensor data to perceive aspects of a surrounding environment of a vehicle. In one embodiment, a method includes, in response to identifying attributes of the surrounding environment from sensor data of one or more sensors, selecting a perception technique from a plurality of perception techniques according to a human-based perception model that correlates the plurality of perception techniques with the attributes to identify which of the plurality of perception techniques efficiently process the sensor data. The method includes analyzing the sensor data using the perception technique to perceive characteristics of the surrounding environment that pertain to autonomously controlling the vehicle. The method includes autonomously controlling the vehicle according to the characteristics to navigate through the surrounding environment.

Abstract: Compositions and process for optimizing oxygen reduction and oxygen evolution reactions are provided. Oxygen reduction and oxygen evolution catalysts include oxide compositions having a general formula a formula A2-xMOy, where x is electrochemically tuned to find optimal A content that delivers the best catalytic performance in a chemical system. The process provides the ability to find the optimal catalytic performance by tuning A and hence, the binding strength of O.

Abstract: System, methods, and other embodiments described herein relate to a device for providing mobility assistance to a user. In one embodiment, a mobility system includes a support component including at least a waist device that is configured to secure the mobility system to the user at a waist area of the user. The mobility system includes a limb attached to the support component and extendable from the support component to a floor when the user is in a standing position. The limb is configured to support the user by providing a rigid structure between the floor and the user. The limb is configured to assist the user in transitioning from a seated position to the standing position by applying a substantially upward force to the user through the support component when transitioning to the standing position.

Abstract: Embodiments described herein disclose methods and systems for object recognition using optimal experimental design. Using detection information from the sensor systems, a detection hypothesis can be generated for the detected object. The detection hypothesis can include 3D models, which have distinctive locations. The distinctive locations can be compared to identified distinctive locations using location estimators. Distinctive locations allow for rejection of a hypothesis, should any distinctive location not have an identified distinctive location on the detected object or within the detection information. In this way, recognition of objects can be performed more quickly and efficiently.

Abstract: A computing system for an autonomous vehicle includes one or more processors for controlling operation of the computing system, and a memory for storing data and program instructions usable by the one or more processors, wherein the one or more processors are configured to execute instructions stored in the memory to: evaluate a first command regime received from a first teleoperator responsive to a predetermined driving situation of the autonomous vehicle; evaluate a second command regime received from a second teleoperator responsive to the predetermined driving situation; responsive to evaluation of the first and second command regimes, select an command regime for implementation by the computing system in response to the predetermined driving situation; and control the autonomous vehicle to implement the selected command regime.

Abstract: Systems, methods and devices for direct conversion of chemical energy into mechanical energy are provided. The system can have a flow of ions, such as between an anode and a cathode. In between the flow of ions, is a membrane, either synthetic or biologically derived. Proteins are bound to that membrane. Further, the proteins are responsive to the ions and undergo a conformational shift, thus using the presence or absence of ions for creating movement. This portion can be referred to as a “kinetic cell”. A portion of the proteins, such as the aqueous portion, can be tethered to a moveable substrate, which translates the motion in the kinetic cell to the exterior. Multiple kinetic cells can be used in coordination to increase the kinetic force generated, analogous to a battery stack.