Abstract: Embodiments described herein provide a supervisor for fault management at a production system. During operation, the supervisor can obtain a set of sensor readings and a state of the production system. A respective sensor reading is an output of a sensor in the production system. The supervisor can then determine, using an artificial intelligence (AI) model, whether the set of sensor readings accommodates a fault associated with a corresponding sensor. Subsequently, the supervisor can determine an action that mitigates an effect of the fault and modify the set of sensor readings based on the action. Here, the modified set of sensor readings is used by a controller that controls the production system.

Abstract: A self-destructing device includes a frangible substrate having at least one pre-weakened area. A heater is thermally coupled to the frangible substrate proximate to or at the pre-weakened area. When activated, the heater generates heat sufficient to initiate self-destruction of the frangible substrate by fractures that propagate from the pre-weakened area and cause the frangible substrate to break into many pieces.

Abstract: A system determines, in a graph which represents a system of components: vulnerability nodes representing known vulnerabilities to the system, including exposed and non-exposed vulnerability nodes associated with an exploitation likelihood; and dependency nodes representing components in the system, including direct and indirect dependency nodes associated with an exposure factor indicating an amount of degradation based on exploitation of an associated vulnerability. The system calculates, across all non-exposed vulnerability nodes and all direct dependency nodes, a score which indicates an attack volume based on at least: a respective second likelihood associated with a non-exposed vulnerability node; an exposure factor associated with a dependency node which represents a component directly degraded based on exploitation of a vulnerability; and a loss of utility of the component.

Abstract: A system determines an on/off feature and vulnerability and dependency nodes in a graph which represents a system of components. The feature enables vulnerability nodes based on a probability that a vulnerability will be exploited, and a vulnerability degrades a utility of one or more components based on an exposure factor. The system calculates, for a path in the graph to a component, a loss of utility of a given dimension of multiple dimensions based on a combiner operator and a logic operator. The combiner operator takes inputs which represent a weighted probability that the given dimension is degraded, and the logic operator defines the inputs based on a probability and exposure factor. The system aggregates calculated losses of utility across the multiple dimensions for the system components. The system selects a combination of possible on/off feature values which results in a lowest loss of utility for the components.

Abstract: A system and method for providing Website navigation recommendations is provided. A Web page of interest is identified as a destination Web page. A domain of Web pages related to the destination Web page is determined. Information is extracted from each Web page in the domain and a recommendation comprising instructions for navigating to the destination Web page is generated based on the extracted information.

Abstract: Methods and systems useful for artificial intelligence-assisted generation of viable hypotheses may have (1) an ability to rapidly enumerate and test a diverse set of mathematically sound and parsimonious physical hypotheses, starting from a few basic assumptions on the embedding spacetime topology; (2) a distinction between non-negotiable mathematical truism (e.g., conservation laws or symmetries), that are directly implied by properties of spacetime, and phenomenological relations (e.g., constitutive laws), whose characterization relies indisputably on empirical observation, justifying targeted use of data-driven methods (e.g., machine learning (ML) or polynomial regression); and (3) a “simple-first” strategy (following Occam’s razor) to search for new hypotheses by incrementally introducing latent variables that are expected to exist based on topological foundations of physics.

Abstract: Systems comprising a carbon fiber reactor for fabricating carbon fiber, the reactor comprising a receptacle for containing a carbon-metal melt, and a plurality of nozzles through which a plurality of menisci are formed by the carbon-metal melt for contact with a carbon seed to fabricate the carbon fiber; and a heater for heating the carbon-metal melt.

Abstract: One embodiment can provide a system for detecting anomaly for high-dimensional sensor data associated with one or more machines. During operation, the system can obtain sensor data from a set of sensor associated with one or machines, generate a first set of outputs by using a set of clustering models learned in parallel from the unlabeled sensor data and user-provided partial label information, generate a second set of outputs by using a set of feed-forward neural network (FNN) models learned in parallel from the first set of outputs and the unlabeled sensor data, and determine whether an anomaly is present in the operation of the one or more machines based on the second set of outputs and a user-specified threshold.

Abstract: A geometry of a substrate surface is received at a neural network. The neural network is trained using one or more training sets. Each training set comprises a different type of substrate geometry and a collection of manufacturing process parameters. The substrate is configured to receive at least one liquid droplet. A shape of the at least one droplet after it has been deposited on the substrate is determined based on the received geometry. An output representing the determined shape of the at least one droplet is produced.

Abstract: One embodiment facilitates generating synthetic data objects using a semi-supervised GAN. During operation, a generator module synthesizes a data object derived from a noise vector and an attribute label. The system passes, to an unsupervised discriminator module, the data object and a set of training objects which are obtained from a training data set. The unsupervised discriminator module calculates: a value indicating a probability that the data object is real; and a latent feature representation of the data object. The system passes the latent feature representation and the attribute label to a supervised discriminator module. The supervised discriminator module calculates a value indicating a probability that the attribute label given the data object is real. The system performs the aforementioned steps iteratively until the generator module produces data objects with a given attribute label which the unsupervised and supervised discriminator modules can no longer identify as fake.

Abstract: Disclosed herein are implementations of a particles-transferring system, particle transferring unit, and method of transferring particles in a pattern. In one implementation, a particles-transferring system includes a first substrate including a first surface to support particles in a pattern, particle transferring unit including an outer surface to be offset from the first surface by a first gap, and second substrate including a second surface to be offset from the outer surface by a second gap. The particle transferring unit removes the particles from the first surface in response to the particles being within the first gap, secures the particles in the pattern to the outer surface, and transports the particles in the pattern. The second substrate removes the particles in the pattern from the particle transferring unit in response to the particles being within the second gap. The particles are to be secured in the pattern to the second surface.

Abstract: A method is provided. The method includes obtaining an enhanced state graph. The enhanced state graph represents a set of objects within an environment and a set of positions of the set of objects. The enhanced state graph includes a set of object nodes, a set of property nodes and a set of goal nodes to represent a set of objectives. The method also includes generating a set of instructions for a set of mechanical systems based on the enhanced state graph. The set of mechanical systems is configured to interact with one or more of the set of objects within the environment. The method further includes operating the set of mechanical systems to achieve the set of objectives based on the set of instructions.

Abstract: Embodiments provide a system and method for extracting configuration-related information for reasoning about the security and functionality of a composed system. During operation, the system determines, by a computing device, information sources associated with hardware and software components of a system, wherein the information sources include at least specification sheets, standard operating procedures, user manuals, and vulnerability databases. The system selects a set of categories of vulnerabilities in a vulnerability database, and ingests the information sources to obtain data in a normalized format. The system extracts, from the ingested information sources, configuration information, vulnerability information, dependency information, and functionality requirements to create a model for the system.

Abstract: A 3D printer includes a nozzle and a camera configured to capture an image, a video, or both of a plurality of drops of liquid metal being jetted through the nozzle. The 3D printer also includes a computing system configured to measure a signal proximate to the nozzle based at least partially upon the image, the video, or both. The computing system is also configured to determine one or more metrics that characterize a behavior of the drops based at least partially upon the signal.

Abstract: A computer-implemented system and method for updating user interest profiles is provided. An interest profile having a mapping of interest categories and interest scores for each interest category is obtained. User generated items are identified from online profiles and a portion of the user generated items are associated with a time stamp. A similarity mapping is performed by comparing each user generated item with each interest category. An interest index score is determined for each interest category based on one of the user generated items compared. A threshold is applied to the interest index scores for each category. The interest score associated with one such category in the mapping is replaced with the interest index score, when the interest index score for that category is above the threshold. The interest profile is adjusted by applying a decay to those interest index scores associated with time stamps determined for the categories.

Abstract: A sensor system includes a sensor network comprising at least one optical fiber having one or more optical sensors. At least one of the optical sensors is arranged to sense vibration of an electrical device and to produce a time variation in light output in response to the vibration. A detector generates an electrical time domain signal in response to the time variation in light output. An analyzer acquires a snapshot frequency component signal which comprises one or more time varying signals of frequency components of the time domain signal over a data acquisition time period. The analyzer detects a condition of the electrical device based on the snapshot frequency component signal.

Abstract: One or more live substances is cultured at a plurality of test locations of a test vessel. The test locations include a thermochromic material and one or more test substances. A spectral shift in light emanating from the thermochromic material of the test locations is detected. The spectral shift occurs in response to an increase or decrease in energy conversion by the live substance. An effect of the one or more test substances on the live substances is determined based on the detected spectral shift.

Abstract: A method of labeling a dataset of input samples for a machine learning task includes selecting a plurality of pre-trained machine learning models that are related to a machine learning task. The method further includes processing a plurality of input data samples through each of the pre-trained models to generate a set of embeddings. The method further includes generating a plurality of clusterings from the set of embeddings. The method further includes analyzing, by a processing device, the plurality of clusterings to extract superclusters. The method further includes assigning pseudo-labels to the input samples based on analysis.

Abstract: A method of three-dimensional manufacturing including depositing a layer of powder in a powder bed, the layer comprising particles of at least a first component of a two-component reactive material, dispensing a first solution containing at least one of a solvent and a second component of the two-component reactive material the solvent selected to cause the two components to form one or more cross-linked regions of a three-dimensional object, iterating the depositing and dispensing to form subsequent layers of the three-dimensional object until the object is formed, and removing any unwanted particles from the object. A three-dimensional manufacturing system has a powder bed to hold layers of particles, and one or more print heads positioned to dispense liquid onto the powder bed to cause the particles for form one or more cross-linked regions in layers of a solid object.

Abstract: A set of load responses of an asset for a sample of traffic loading events caused by objects of unknown weight is measured. At least one statistical parameter is determined from the set of load responses. A corresponding statistical parameter of known object weights loading the asset is determined. An object weight is assigned to a load response of the set of load responses based on correlation of the extracted statistical parameter to the corresponding statistical parameter.