Abstract: A beam accelerator system operable to produce a medical isotope, including an ion accelerator that generates an ion beam; a low-pressure chamber; an anode adjacent and fluidly connected to the low-pressure chamber; a plasma window adjacent and fluidly connected to the anode; and a cathode housing adjacent and fluidly connected to the plasma window. The plasma window has a plurality of plates, each plate having an aperture that is aligned with an aperture in one or more adjacent plates to form a plasma channel. One or more plates in the plurality of plates includes a unitary plate having an aperture therein, and one or more cooling channels entering the unitary plate at a first side of the unitary plate and exiting the unitary plate at a second side of the unitary plate. The one or more cooling channels run through a thickness of the unitary plate.

Abstract: Securing package delivery via unmanned aerial vehicles (UAVs), drones, robots, etc. is contemplated. The package delivery may include assigning a provenance certificate to a package scheduled for drone delivery, determining a plurality of vehicles to be used in transporting the package to an autonomous receiver, determining credentials assigned to each of the vehicles, pairing the credentials with the package to generate a delivery pairing, the delivery pairing identifying the credentials of the vehicles authorized to transport the package, and transporting the delivery pairing to each of the vehicles to prevent vehicles having certificates omitted from the delivery pairing from transporting the package.

Abstract: Some embodiments described herein are directed to a beverage container including a drinking opening, a closure, and a seal. The closure is configured to selectively close the drinking opening. The seal is coupled to the closure and configured to seal the drinking opening when the closure closes the drinking opening. The closure includes a shaft extending down from an interior of the closure. The seal covers the shaft, and a slit extends through a portion of the seal covering the shaft.

Abstract: A positive displacement fluid pump is provided. The fluid pump includes a housing defining an internal cavity. A motor having a drive shaft that rotates about an axis is housed within the internal cavity of the housing. An internal plate is adjacent the motor and includes a central bore through which the drive shaft extends. The fluid pump further includes an external plate including an inlet in fluid communication with a suction port and an outlet in fluid communication with a delivery port. A pumping ring is sandwiched between the internal and external plates, and a pumping arrangement is located within the pumping ring and axially between the internal plate and the external plate. The pumping arrangement is rotatably coupled to the drive shaft such that rotation of the pumping arrangement by the drive shaft causes fluid to be pumped from the suction port to the delivery port.

Abstract: A method for registering a handler in a configured sequence of handlers includes: receiving, by a processor and memory implementing a sequencer infrastructure, a configuration script defining a sequence including one or more handler identifiers and corresponding one or more handler arguments; invoking a handler initialization entry point for a handler corresponding to a handler identifier, the invoking the handler initialization entry point including: selecting, based on the corresponding one or more handler arguments, a handler type from among two or more of: a configured instance of the handler; a configured alternate handler different from the handler; an alternate sequence of handlers; an expression; and a no-op handler; and initializing a configured handler based on the handler type selected and the corresponding one or more handler arguments; and adding the configured handler to a configured sequence of handlers.

Abstract: Example aspects of the present disclosure are directed to systems and methods that learn a compressed representation of a machine-learned model (e.g., neural network) via representation of the model parameters within a reparameterization space during training of the model. In particular, the present disclosure describes an end-to-end model weight compression approach that employs a latent-variable data compression method. The model parameters (e.g., weights and biases) are represented in a “latent” or “reparameterization” space, amounting to a reparameterization. In some implementations, this space can be equipped with a learned probability model, which is used first to impose an entropy penalty on the parameter representation during training, and second to compress the representation using arithmetic coding after training. The proposed approach can thus maximize accuracy and model compressibility jointly, in an end-to-end fashion, with the rate-error trade-off specified by a hyperparameter.

Abstract: Disclosed herein is a wireless mesh network comprised of ultra-high-capacity nodes that are capable of establishing ultra-high-capacity links (e.g., point-to-point or point-to-multipoint bi-directional communication links) using a millimeter wave spectrum, including but not limited to 28 Ghz, 39 Ghz, 37/42 Ghz, 60 Ghz (including V band), or E-band frequencies, as examples. The higher capacity and/or extended range of these ultra-high-capacity nodes/links may be achieved via various advanced signal processing techniques. Further, these ultra-high-capacity nodes/links may be used in conjunction with other types of point-to-point and/or point-to-multipoint links to build a multi-layer wireless mesh network.

Abstract: Static base station structures having a plurality of detachable mobile components, systems, and methods of receiving, storing, and charging the plurality of detachable mobile components are disclosed herein. The static base station structure may be a desk or kiosk including an electronic device. The plurality of detachable mobile components may be wheeled mobile desks or kiosks, each having their own electronic device, and detachable from and independently operable from the static structure. The electronic devices of the plurality of detachable mobile components may maintain wireless communications with the electronic device of the static structure, or with the electronic device of another remote device, even when physically separated. When the detachable mobile component is no longer needed, it may be returned to a docking station within the static base station structure for storing and charging.

Abstract: A centerbody assembly for a pressure regulator includes: a centerbody having a process surface, at least one outlet orifice disposed in the centerbody; an outlet port positioned in fluid communication with the at least one outlet orifice; an inlet port; a valve chamber positioned in the centerbody in fluid communication with the inlet port, the valve chamber including a valve seat positioned in fluid communication with the process surface; a spherical valve element disposed in the valve chamber and moveable between an open position in which it is spaced away from the valve seat, and a closed position in which it seals against the valve seat; a flexible control diaphragm facing the process surface, wherein a perimeter of the control diaphragm is bonded to the centerbody so as to define a seal that blocks the passage of fluid; and wherein the valve element is mechanically interconnected with the control diaphragm.

Abstract: A composition including at least about 50 wt % basalt and no more than about 20 wt % binding agent, wherein the composition is in the form of a plurality of particles. The composition can find applications in agriculture, horticulture, and gardening, and can be used. e.g., as soil amendment.

Abstract: There is included a method and apparatus comprising computer code for instance-wise adaptive knowledge injection in a pre-trained language model (PTLM) including determining a necessity of external knowledge in a plurality of queries of a first dataset based on a likelihood that a respective query is solved by internal knowledge of a target model. Then, the one or more queries determined to need external knowledge may be augmented with pieces of external knowledge. A combined dataset may be generated by combining the first dataset and the one or more augmented queries, and the combined dataset may be applied to the target model.

Abstract: Some embodiments of the invention provide a method for WAN (wide area network) optimization for a WAN that connects multiple sites, each of which has at least one router. At a gateway router deployed to a public cloud, the method receives from at least two routers at least two sites, multiple data streams destined for a particular centralized datacenter. The method performs a WAN optimization operation to aggregate the multiple streams into one outbound stream that is WAN optimized for forwarding to the particular centralized datacenter. The method then forwards the WAN-optimized data stream to the particular centralized datacenter.

Abstract: Methods and apparatus for a desaturation circuit having a temperature compensated voltage threshold for protecting a power transistor, such as a FET. A comparator module has a first input compared to a voltage source associated with current through the FET and a second input coupled to the voltage threshold circuit. As a resistance of the FET changes due to temperature changes, the voltage threshold changes to compensate for the resistance change of the FET.

Abstract: A method of writing voxels to a substrate using a laser writing system comprises forming a first voxel at a first position in a substrate using a first laser pulse; detecting light emitted or scattered by the substrate as a result of forming the first voxel; determining whether the detected light satisfies a predetermined constraint; and, when the detected light does not satisfy the predetermined constraint, adjusting an amplitude of a second laser pulse. Light emission or scattering from the substrate as a result of forming a voxel is related to the properties of the formed voxel. By monitoring such emission or scattering, it is made possible to compensate for variations in performance of the laser writing system. Also provided herein are a laser writing system and computer program product which implement the method.

Abstract: Systems and methods are provided for a battery management system (BMS) having a protection circuit. In one example, a vehicle battery system may include the BMS, the BMS including a cutoff circuit coupled to a short-circuit protection circuit, and a battery pack, wherein the short-circuit protection circuit may include a diode array, cathodes of the diode array being coupled to a positive terminal post of the battery pack and anodes of the diode array being coupled to a negative terminal post of the battery pack. In some examples, the cutoff circuit may further be coupled to a reverse bias protection circuit including a switchable current path arranged between a control input of the cutoff circuit and an output of the cutoff circuit. In this way, the vehicle battery system may be protected from unexpected voltage conditions via the BMS redirecting and dissipating excess current away from the cutoff circuit.

Abstract: An optical power and data delivery system includes an optical base station and an autonomous wireless hotspot device. The optical base station includes routers, a media converter device, a optical data transceiver configured to transmit the received data laser light beams; one or more power laser devices configured to transmit a plurality of power laser light beams, a beam combining device to generate and transmit combined power and data laser light beams through a first optical antenna. The autonomous wireless hotspot device includes a second optical antenna configured to receive the combined data and power laser light beams, an optical dividing device, a optical data transceiver, a media converter device and wireless communication transceivers to transmit wireless data signals and one or more laser power converters to receive the power laser light beams and to convert the power laser light beams into electrical energy.

Abstract: A data processing system implements a receiving a textual input comprising a query for a first image. The data processing system also implements analyzing the textual input to determine a predicted color palette associated with a subject matter of the query; and procedurally generating the first image using the predicted color palette. Another implementation of the data processing system implements providing the textual input to a first machine learning model to obtain the first image, the first machine learning model being trained using a dataset comprising abstract imagery and analyzing the textual input using the first machine learning model to obtain the first image in response to receiving the textual input.

Abstract: This application relates to sensor systems and related devices and methods. A sensor device can include an image sensor configured to generate image data based on infrared light received at the image sensor, a primary lens positioned to direct light from a scene onto the image sensor, and an array of microlenses positioned between the image sensor and the primary lens, each microlens configured to focus the light from the scene onto a corresponding portion of the image sensor. A method can include receiving plenoptic image data based on infrared light, generating a two-dimensional image of a scene, and determining a depth to an object represented by the pixel based on the plenoptic image data.