Abstract: A system and method for robust estimation of state parameters from internal readings in a sequence of images are provided. Various techniques can be implemented to address observation noise and/or underlying process noise to stabilize the readings.

Abstract: A system captures, by a recording device, a scene with physical objects, the scene displayed as a three-dimensional (3D) mesh. The system marks 3D annotations for a physical object and identifies a mask. The mask indicates background pixels corresponding to a region behind the physical object. Each background pixel is associated with a value. The system captures a plurality of images of the scene with varying features, wherein a respective image includes: two-dimensional (2D) projections corresponding to the marked 3D annotations for the physical object; and the mask based on the associated value for each background pixel. The system updates the value of each background pixel with a new value. The system trains a machine model using the respective image as generated labeled data, thereby obtaining the generated labeled data in an automated manner based on a minimal amount of marked annotations.

Abstract: A system and method for reading digits using VGG-16 backbone are provided to create visual features followed by two layers of non-linear fully connected units which are then fed to 8 categorical symbol units and a single linear length unit. The 8 categorical units provide an ordered representation of the numerical reading with required punctuation such as decimal points or colons. Training on synthetic digits followed by augmentations to create a robust detector are implemented without the need for real-world training data.

Abstract: An augmented data set is generated by combining chroma key substitutions for background and occlusion augmentations for foreground to enable training of object detectors or action recognition that transfers to multiple environments and is robust to occlusion due to other objects or human hands.

Abstract: A system and a method for secure control of a physical system are described. During operation, the system can obtain measurement of one or more sensors associated with the physical system. The system then estimates a state of the physical system based on the measurement of the sensors. Subsequently, the system generates a feedback control signal based on the estimated state and generates a watermarking signal based on a stored estimated state of the physical system at a previous time instant. The system then generates a watermarked control signal by combining the feedback control signal and the watermarking signal and applies the watermarked control signal to the physical system to regulate the state of the physical system, thereby facilitating secure control of the physical system.

Abstract: An approach to automating hybrid manufacturing is provided. This approach first splits an item of manufacture into irreducible pieces called atoms, then extracts constraints between the atoms, and then uses a planner to find the best way to assemble the atoms into the item of manufacture given the constraints. The approach splits the item into atoms based on various properties such as the presence or absence of support, the need for scaffolding, tool reachability constraints, and the presence of cycles of dependencies which prevent the item from being manufactured.

Abstract: A modification of a topology optimization formulation for part-level generation is provided such that it includes additional constraints that ensure that the inertial quantities (e.g., COM, Mol, and mass) of an automatically generated part are equivalent to specified targets. These targets are the inertial values used in an initial assembly-level simulation. The reaction and external forces acting on the part of interest resulting from assembly-level simulation provide the boundary conditions used for part-level generation. By constraining the inertial properties in this way, there is no need to evaluate the assembly-level dynamics each time a part is generated.

Abstract: One embodiment can provide a method and system for estimating a remote quantity of interest (QoI). During operation, the system can receive, over a communication channel, a radio frequency (RF) signal carrying an estimate of the QoI measured by a sensor. The system can estimate probability distributions of a set of random channel parameters associated with the HF communication channel. The system can further reconstruct the estimate based on the probability distributions of the channel parameters and the received RF signal, determine a level of uncertainty associated with the reconstructed estimate, and combine reconstructed estimates from multiple sensors based on the determined level of uncertainty associated with each reconstructed estimate to output a combined estimate of the QoI.

Abstract: Methods comprising: evaporating a catalyst source to produce a catalyst gas; condensing the catalyst gas to produce a catalyst vapor comprising catalyst droplets suspended in a gas phase; and contacting the catalyst vapor with a hydrocarbon gas to catalyze a decomposition reaction of the hydrocarbon gas into hydrogen gas and carbon. And, systems comprising: a catalyst source evaporator that provides a first stream to a reactor; a hydrocarbon source that provides a second stream to the reactor; a cooling column coupled to the reactor via a third stream comprising hydrogen, catalyst liquid, solid carbon, optionally catalyst gas, and optionally unreacted hydrocarbon gas such that the cooling column receives the third stream from the reactor; and wherein the cooling column has effluent streams that include (a) a fourth stream that comprises hydrogen and optionally catalyst gas and (b) a fifth stream that comprises catalyst liquid.

Abstract: One embodiment of the present invention provides a system that recommends activities. During operation, the system receives a piece of content obtained from text or converted to text from speech. The system then analyzes the received content to identify any activity type, indication of willingness to participate in any type of activities, and at least one piece of temporal information, which can be implicitly and/or explicitly stated in the content, and/or one piece of location information associated with the activity type. The system further recommends one or more activities, venues, and/or services that afford or support activities for a user based on the information extracted from the content.

Abstract: One embodiment provides a method for automated design of a physical system. The method can include obtaining design requirements associated with the physical system and iteratively performing, by a computer, a reinforcement learning (RL) process and a nonlinear optimization process to generate a design solution. The RL process can generate a topology represented as a model of the physical system using a modeling language. The generated topology can specify a number of components and connections among the components. The nonlinear optimization process can determine parameters of the components in the topology based on the model and a loss function. The method can further include outputting the design solution of the physical system based on the generated topology and the determined parameters of the components, thereby facilitating construction of the physical system.

Abstract: Embodiments described herein provide a system for facilitating efficient troubleshooting for a product. During operation, the system can identify an artificial-intelligence- (AI-) based dialog model operating based on a structured representation, which can indicate sequential troubleshooting steps to be performed by a user. The system can provide a machine utterance of the dialog model corresponding to a troubleshooting step to the user. The system can then search the structured representation for a semantic match for a user utterance obtained in accordance with the dialog model from the user. If the semantic match indicates an anticipated option associated with the machine utterance, the system can traverse a current branch of the structured representation using the dialog model based on the anticipated option. Otherwise, if the semantic match indicates an option on a different branch, the system can jump to the option on the different branch for subsequent traversal.

Abstract: System and methods for relational time-series learning are provided. Unlike traditional time series forecasting techniques, which assume either complete time series independence or complete dependence, the disclosed system and method allow time series forecasting that can be performed on multivariate time series represented as vertices in graphs with arbitrary structures and predicting a future classification for data items represented by one of nodes in the graph. The system and methods also utilize non-relational, relational, temporal data for classification, and allow using fast and parallel classification techniques with linear speedups. The system and methods are well-suited for processing data in a streaming or online setting and naturally handle training data with skewed or unbalanced class labels.

Abstract: One embodiment provides a method and a system for reconstructing symbols transmitted over a high frequency (HF) communication channel. During operation, the system can receive, at a receiver, a radio frequency (RF) signal carrying an input data frame and transmitted over the HF communication channel. The input data frame includes a number of known symbols followed by a number of unknown symbols. The system can determine a set of channel parameters associated with the HF communication channel based on the received RF signal and the known symbols and reconstruct, using a machine-learning technique, the unknown symbols based on the determined channel parameters and the received RF signal.

Abstract: A transferable hybrid method for prognostics of engineering systems based on fundamental degradation modes is provided. The method includes developing a degradation model that represents degradation modes shared in different domains of application through the integration of physics and machine learning. The system measures sensor signals and data processing provides for extracting health indicators correlated with the fundamental degradation modes from sensors data. For the integration of physics and machine learning, the degradation mode is separated into different phases. Before the accelerated degradation phase of a system, the method is looking out to detect when the accelerated phase begins. When accelerated phase is active, the system applies a machine-learning model to provide information on the accelerated degradation phase, and evolves the degradation towards a failure threshold in a simulation of the updated physics-based model to predict the degradation progression.

Abstract: One embodiment provides a system which facilitates construction of an ensemble of neural network-based classifiers that optimize a diversity metric. During operation, the system defines a diversity metric based on pairwise angles between decision boundaries of three or more affine classifiers. The system includes the diversity metric as a regularization term in a loss function optimization for designing a pair of mutually orthogonal affine classifiers of the three or more affine classifiers. The system trains one or more neural networks such that parameters of the one or more neural networks are consistent with parameters of the affine classifiers to obtain an ensemble of neural network-based classifiers which optimize the diversity metric. The system predicts an outcome for a testing data object based on the obtained ensemble of neural-network based classifiers which optimize the diversity metric.

Abstract: One embodiment provides a method and system which facilitates optimizing a pair of affine classifiers based on a diversity metric. During operation, the system defines a diversity metric based on an angle between decision boundaries of a pair of affine classifiers. The system includes the diversity metric as a regularization term in a loss function optimization for designing the pair of affine classifiers, wherein the designed pair of affine classifiers are mutually orthogonal. The system predicts an outcome for a testing data object based on the designed pair of mutually orthogonal affine classifiers.

Abstract: A system determines an input video and a first annotated image from the input video which identifies an object of interest. The system initiates a tracker based on the first annotated image and the input video. The tracker generates, based on the first annotated image and the input video, information including: a sliding window for false positives; a first set of unlabeled images from the input video; and at least two images with corresponding labeled states. A semi-supervised classifier classifies, based on the information, the first set of unlabeled images from the input video. If a first unlabeled image is classified as a false positive, the system reinitiates the tracker based on a second annotated image occurring in a frame prior to a frame with the false positive. The system generates an output video comprising the input video displayed with tracking on the object of interest.

Abstract: Embodiments described herein provide a design architecture for co-designing a controller and a watermarking signal for a cyber-physical system. During operation, the architecture can determine, in conjunction with each other, respective values of a first set of parameters indicating operations of the controller and a second set of parameters representing the watermarking signal. Here, the watermarking signal is combinable with a control signal from the controller for monitoring an output signal of the cyber-physical system for detecting malicious data at different time instances. Subsequently, the architecture can determine a state manager for determining the states of the cyber-physical system from the monitored output signal based on the first and second sets of parameters. The architecture can also determine a detector capable of identifying presence of an attack from the states of the cyber-physical system at a plurality of time instances using the watermarking signal.

Abstract: A system and method provide a combination of a modular message structure, a priority-based message packing scheme, and a data packet queue management system to optimize the information content of a transmitted message in, for example, the Ocean of Things (OoT) environment. The modular message structure starts with a header that provides critical information and reference points for time and location. The rest of the message is composed of modular data packets, each of which has a data ID section that the message decoder uses for reference when reconstructing the message contents, an optional size section that specifies the length of the following data section if it can contain data of variable length, and a data section that can be compressed in a manner unique to that data type.