Abstract: Select an initial weight vector for a convex optimization sub-problem associated with a neural network having a non-convex network architecture loss surface. With at least one processor, approximate a solution to the convex optimization sub-problem that obtains a search direction, to learn a common classifier from training data. With the at least one processor, update the initial weight vector by subtracting the approximate solution to the convex optimization sub-problem times a first learning rate. With the at least one processor, repeat the approximating and updating steps, for a plurality of iterations, with the updated weight vector from a given one of the iterations taken as the initial weight vector for a next one of the iterations, to obtain a final weight vector for the neural network, until convergence to a global minimum is achieved, to implement the common classifier.

Abstract: In aspects of the disclosure, a method comprises training, by a computing system, a dynamics model of a cooperative multi-agent reinforcement learning (c-MARL) environment. The method further comprises processing, by the computing system, a perturbation optimizer to generate a state perturbation of the c-MARL environment, based on the dynamics model. The method further comprises selecting one or more agents of the c-MARL system as having enhanced vulnerability. The method further comprises attacking, by the computing system, the c-MARL system based on the state perturbation and the selected one or more agents.

Abstract: A contact lens for shaping a cornea of an eye of a patient to treat high cylinder astigmatism may include a peripheral portion, an alignment portion, a treatment portion, a reverse portion, and a staining feature. The peripheral portion may cause migration of epithelial cells from a periphery of a cornea of the eye of the patient towards a center of the cornea of the eye of the patient. The alignment portion may align the contact lens on the cornea. The treatment portion may cause migration of the epithelial cells from the center of the cornea towards the periphery of the cornea. The reverse portion may relieve pressure caused by the migration of the epithelial cells. The staining feature may contact the cornea, thereby reshaping the cornea to treat high cylinder astigmatism. The staining feature may apply a ribbon-shaped staining pattern to the cornea across the contact lens.

Abstract: In example aspects of this disclosure, a method includes generating, by one or more computing devices, a parametric model that expresses condition states for each of a plurality of assets, and the probability of the assets transitioning between the condition states; generating, by the one or more computing devices, stochastic degradation predictions of a group of the assets, based on the condition states and the probability of transitioning between the condition states for at least some of the assets; and generating, by the one or more computing devices, a maintenance schedule based on: the stochastic degradation predictions of the group of the assets, costs of corrective maintenance for assets in a failed state, and costs of scheduled maintenance for the assets.

Abstract: Methods, systems, and computer program products for a decision-improvement framework are provided herein. A computer-implemented method includes obtaining regression functions that predict an output of processes of a physical system based on inputs received at each process; automatically generating one or more constraints and one or more objective functions for a model for the physical system based on the regression functions and a representation of the physical system, where the representation specifies relationships between at least a portion of the processes; identifying a set of parameter values for controlling the physical system based on the model; generating a score, for the set of parameter values, based on a predicted improvement of the physical system relative to historical performance of the physical system; and in response to the generated score satisfying a threshold, causing the physical system to be configured in accordance with the set of parameter values.

Abstract: An elevator of a medical device may comprise: a plurality of segments joined together by at least one hinge, and a control element coupled to at least one segment of the plurality of segments. A first segment of the plurality of segments may be a discrete structure. Proximal or distal movement of the control element may be configured to move the at least one segment of the plurality of segments relative to another of the plurality of segments.

Abstract: A system and method for mind-body relaxation may include a plurality of vibration modules, a light-providing member, a controller, a relaxation substrate, a pillow member, a graphical user interface (GUI), and network connection circuitry.

Abstract: A computer implemented method for certifying robustness of image classification in a neural network is provided. The method includes initializing a neural network model. The neural network model includes a problem space and a decision boundary. A processor receives a data set of images, image labels, and a perturbation schedule. Images are drawn from the data set in the problem space. A distance from the decision boundary is determined for the images in the problem space. A re-weighting value is applied to the images. A modified perturbation magnitude is applied to the images. A total loss function for the images in the problem space is determined using the re-weighting value. A confidence level of the classification of the images in the data set is evaluated for certifiable robustness.

Abstract: Devices, systems, and methods for treating vascular defects are disclosed herein. One aspect of the present technology, for example, includes an occlusive device comprising a mesh having a low-profile state for intravascular delivery to the aneurysm and a deployed state, the mesh comprising a first end portion, a second end portion, and a length extending between the first and second end portions, and a first lateral edge, a second lateral edge, and a width extending between the first and second lateral edges. The mesh may have a predetermined shape in the deployed state in which (a) the mesh is curved along its width, (b) the mesh is curved along its length, and (c) the mesh has an undulating contour across at least a portion of one or both of its length or its width. The mesh is configured to be positioned within the aneurysm in the deployed state such that the mesh extends over the neck of the aneurysm.

Abstract: A handle of an endoscope may comprise a body portion that defines a first chamber having a first feature configured to support at least a first component of a steering assembly, and a second chamber having a second feature configured to support at least one fluidic component or electronic component.

Abstract: Bias supplies and bias control methods are disclosed. One method comprises applying an asymmetric periodic voltage waveform and providing a corresponding current waveform at an output node relative to a return node; receiving a signal to change from a current state of the asymmetric periodic voltage waveform to a next state of the asymmetric periodic voltage waveform; and adjusting, during a transition from the current state to the next state, at least one portion of the asymmetric periodic voltage waveform and simultaneously adjusting a fundamental frequency of the asymmetric periodic voltage waveform to settle at the next state.

Abstract: An extraction cleaner includes a portable extractor and a base adapted to be moved to perform a cleaning operation on a floor surface when the portable extractor is operably mounted to the base. The portable extractor includes a recovery tank assembly provided on the housing adjacent to the recovery tank assembly.

Abstract: A surface cleaning apparatus having a suction source defining a portion of a working air path and a fluid delivery system includes at least one fluid distributor, a fluid supply path, a first pump in the fluid supply path, and a second pump in the fluid supply path. The pumps are operable to effect dispensing cleaning fluid at different volumetric flow rates.

Abstract: Various embodiments relate to convolutional neural networks (CNN). CNN may be provided with a convolution kernel configured with binary weights. The CNN may be trained with the convolution kernel to determine a set of binary weights for the convolution kernel. The set of binary weights may be used for inference of the CNN. Devices, methods, and computer programs are disclosed.

Abstract: Bias supplies, plasma processing systems, and associated methods are disclosed. One bias supply comprises a first inductor coupled between a first node of a switch and an output node where a first node of a second inductor is coupled to one of the output node or the first node of the switch. A voltage source is coupled between a second node of the switch and a second node of the second inductor. A connection is made between the return node and one of the second node of the switch and the second node of the second inductor. The bias supply also comprises a controller configured to cause an application of the periodic voltage between the output node and the return node by repeatedly closing the switch so current through the switch completes a full cycle.

Abstract: Bias supplies, plasma processing systems, and associated methods are disclosed. One bias supply comprises a first inductor coupled between a first node of a switch and an output node where a first node of a second inductor is coupled to one of the output node or the first node of the switch. A voltage source is coupled between a second node of the switch and a second node of the second inductor. A connection is made between the return node and one of the second node of the switch and the second node of the second inductor. The bias supply also comprises a controller configured to cause an application of the periodic voltage between the output node and the return node by repeatedly closing the switch so current through the switch completes a full cycle.

Abstract: An index sequence specifying an index of training data corresponding to a component of a cost function is generated. A first model parameter in the set of model parameters is set to an initial value. Using the index sequence, a neural network model comprising a set of weights is trained. As part of the training, using the index sequence, a learning rate, and a set of gradients, a subset of the set of model parameters is updated. As part of the training, a momentum term is set. As part of the training, using the momentum term as the first model parameter, the updating and the setting are repeated until reaching a training completion condition. The trained neural network model is used to predict an outcome by analyzing live data.

Abstract: A production plant and a method for production of a synthetic gasoline product from a synthetic hydrocarbon mixture produced by post-treatment of a synthetic gasoline product including less than a specified concentration of olefins, such as 6 vol % or 11 vol % from a first synthetic hydrocarbon mixture produced from a mixture of reactive oxygenates, the first synthetic hydrocarbon mixture having T90 of less than 140° C. and including at least the specified concentration of olefins and a second synthetic hydrocarbon mixture, produced from a mixture of reactive oxygenates, the second synthetic hydrocarbon mixture having T90 of more than 150° C., the method including: a. directing the second synthetic hydrocarbon mixture to contact a material catalytically active in hydrocracking under effective hydrocracking conditions, to provide a hydrocracked second synthetic hydrocarbon mixture, b.