Abstract: Neural network systems are provided that comprise one or more neural networks. The first neural network can comprise a convolutional neural network (CNN) long short-term memory (LSTM) architecture for receiving a primary data set comprising text messages and output a primary data structure comprising a text pattern-based feature. The second neural network can comprise a CNN architecture for receiving a secondary data sets derived from the primary data set and output a plurality of secondary data structures. The third neural network can combine the data structures to produce a combined data structure, and then process it to produce a categorized data structure comprising the text messages assigned to targets. The primary data set can comprise hate speech and the categorized data structure can comprise target categories, for example, hate targets. Methods of operating neural network systems and computer program products for performing such methods are also provided.

Abstract: A system and a process for building, monitoring, and rebuilding machine learning models includes building a plurality of machine learning models from a design specification, deploy the plurality of machine learning models for operation, and designating a champion and at least one challenger from the plurality of machine learning models built. The system and the process evaluate the plurality of machine learning models and provide results of the evaluation.

Abstract: Systems, methods, and computer-readable storage media for monitoring risk levels in continuing to allow a potentially compromised payment card to continue to be used. An exemplary system can include a processor which receives a notification of unauthorized use of a card, retrieves a transaction history of authorized use of the card, and performs predictive modeling based on the transaction history based to determine a predicted transaction amount for future transactions. The system can then determine a risk level of continued use of the card and modify processing of the ongoing transaction based on the risk level. The system can then issue a risk level notification to proper authorities.

Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, perform certain acts. The acts can include obtaining reorder likelihood scores for items that a user has ordered historically. The acts also can include grouping the items into groups using a taxonomy. The acts additionally can include adjusting the groups based on a respective number of items in each of the groups and a respective group-specific threshold for each of the groups. The acts further can include ranking the items within the groups based on the reorder likelihood scores. Other embodiments are described.

Abstract: A method for denoising video content includes identifying a first frame block associated with a first frame of the video content. The method also includes estimating a first noise model that represents characteristics of the first frame block. The method also includes identifying at least one frame block adjacent to the first frame block. The method also includes generating a second noise model that represents characteristics of the at least one frame block adjacent to the first frame block by adjusting the first noise model based on at least one characteristic of the at least one frame block adjacent to the first frame block. The method also includes denoising the at least one frame block adjacent to the first frame block using the second noise model.

Abstract: An imaging device may have an array of image sensor pixels arranged in rows and columns and column readout circuitry coupled to the array. The rows of pixels may receive drive signals from row driver circuitry, and the drive signals may be sent from timing circuitry based on the locations of rows within the array. In particular, rows closer to the readout circuitry may require less settling time and therefore be driven faster than the rows further from the readout circuitry. All of the rows may be driven in a single direction, or the array of pixels may have a cut, in which case rows above the cut may be driven up and rows below the cut may be driven down. A frame buffer may be used to store the signals generated by the rows of pixels and may account for the asynchronous read out of image data.

Abstract: Systems for performing an automated, in situ calibration of one or more sensors of internet of things (IoT) systems include one or more emulators capable of generating calibration set points that are applied to the sensors during the calibration process. The systems also include one or more computing devices configured to store the data necessary for the calibrations. The computing devices are further configured to monitor the sensor outputs during normal operation of the IoT systems to check for a loss of calibration or compromised data integrity; execute an automated calibration of a upon the detection of a loss of calibration or data-integrity issue; and validate the calibration results.

Abstract: An intraosseous access system can include a hub coupled to a cannula that can be introduced into a bone of a patient. The system further includes a securement device that can couple with the hub when transitioned from an open orientation to a closed orientation. The securement device includes a first segment and a second segment that are spaced apart from each other when the securement device is in the open orientation and are approximated and secured to each other when the securement device is in the closed orientation. The first segment includes a first receptacle that receives a portion of the hub therein and contacts the hub to restrain movement of the hub relative to the securement device when the securement device is coupled with the hub in the closed orientation. The first segment also includes a first arm coupled to the first receptacle, the first arm biasing the first receptacle toward the hub when the securement device is coupled with the hub in the closed orientation.

Abstract: A method for producing a target including halftone printing an ink, by a printer, onto at least one ultraviolet reflective substrate. The method further includes applying at least one overlay material, by at least one overlay device, onto the at least one ultraviolet reflective substrate such that a photorealistic visible image and a photorealistic non-visible image are dually formed on the at least one ultraviolet reflective substrate.

Abstract: A system for modifying a talocalcaneal spatial relationship in a foot in a body, including a subtalar spacer having an articulating surface and implanted in the sinus tarsi of the foot and a fastener that fastens the subtalar spacer to a bone surface in the foot, in which the articulating surface of the subtalar spacer directs relative movement between the calcaneus and the talus of the foot.

Abstract: A multiple hitch assembly and method of use thereof including a hitch platform having a channel member and multiple hitch member having a block base. A dampening member may be affixed such that coupling of the multiple hitch member by locked engagement of received through holes of the block base and paired side holes disposed in flanges of the channel member compresses the dampening member upon the recessed column of the hitch platform to absorb and thus attenuate the load and associated torque imparted by movement and rub between the hitch platform and multiple hitch member to thereby reduce associated rattle noise, and vibration.

Abstract: Extended release pyridostigmine dosage forms, suitable for maintaining stable plasma concentrations with reduced or minimized initial burst release/dose dumping of pyridostigmine, are provided. The dosage forms include matrix tablets, gastroretentive tablets, and pellets, the latter being suitable for dosing in capsules, tablets, and sachets, as well as for sprinkling on foodstuffs. The disclosure also provides methods for improving patient compliance by administering once-a-day extended release pyridostigmine bromide dosage forms that provide a superior controlled drug release.

Abstract: A connector for use in a free-standing connector port for mating with an external pluggable module is disclosed. The connector has terminals that extend lengthwise of the connector so that cables may be terminated to the terminals and the terminals and cable generally are horizontally aligned together. The connector defines a card-receiving slot and is position in a cage that defines the connector port. The cables exit from the rear of the connector and from the connector port. The connector includes a heat sink that is configured to cool a module inserted into the connector port.

Abstract: A process and system provides for atraumatic preparation of morselized Tissue Particles (TP)s, such as Full Thickness Skin Graft Particles (FTSGPs), cartilage particles and other organ tissue particles, in a liquid medium. The resultant tissue product may be a suspension of Tissue Particles in an aqueous solution and containing highly viable cells and may be rapidly prepared at bedside or in the operating room and conveniently delivered to a patient through a syringe or similar applicator. The morselized Tissues Particles may be used for surgical applications including wound healing, cosmetic surgery, and orthopedic cartilage repairs.

Abstract: An ionization device may be configured to be portable, and to rest on a surface such as a floor or desk top. The ionization device includes an air-intake port, an ion generator, an ozone catalyst for removing at least some ozone from air, and an air discharge. Air enters the device through the air-intake port, and at least some of the air is ionized to remove particulates. The air is then moved past or through the ozone catalyst to remove at least some of the ozone from the air. A controller may be used to monitor particulates, temperature, humidity, and/or other relevant factors and/or to adjust the ionization level.

Abstract: A lock node for storing data and a protected storage unit. The lock node includes an input section which provides a plurality of key maps, each corresponding to one of a plurality of primary keys, respectively, applied to the input section, each key map including at least one main key, a variable lock section producing a derived key from a logical operation on the main keys corresponding to the primary keys applied to the input section, and an output section producing the data in response to the derived key.

Abstract: Described herein are embodiments for end-to-end reinforcement learning based coreference resolution models to directly optimize coreference evaluation metrics. Embodiments of a reinforced policy gradient model are disclosed to incorporate reward associated with a sequence of coreference linking actions. Furthermore, maximum entropy regularization may be used for adequate exploration to prevent a model embodiment from prematurely converging to a bad local optimum. Experiments on datasets compared with state-of-the-art methods verified the effectiveness of embodiments.

Abstract: An underlayment layer is configured to support an artificial turf assembly. The underlayment layer comprises a core with a top side and a bottom side. The top side has a plurality of spaced apart, upwardly oriented projections that define channels suitable for fluid flow along the top side of the core when the underlayment layer is positioned beneath an overlying artificial turf assembly.

Abstract: A fluid flow diverter is provided that includes a diverter body having four ports, a rotating plenum located within the diverter body, and a purge fluid assembly that supplies a purge fluid to the plenum. The plenum has two stop positions that each define a fluid flow path through the diverter. In the first fluid flow path, a first fluid stream goes between the first and second ports, and a second fluid stream goes between the fourth and third ports. In the second flow path, a first fluid stream goes between the first and third ports, and a second fluid stream goes between the fourth and second ports. The purge fluid supplied to the plenum creates a positive pressure fluid barrier that prevents or minimizes cross-contamination of the two fluid streams through the diverter. Also provided is a regenerative thermal oxidizer that includes such a fluid flow diverter.

Abstract: A system and a method can receive a first dataset having a first label and a first context. The system and the method can also generate, at the trained deep neural network, a second dataset having the first label and a second context according to a mapping, wherein a first mapping of the plurality of mapping comprises one or more weights of the trained deep neural network that maps data having the first label and the first context to data having a second label and the first context and a second mapping of the plurality of mapping comprises one or more weights of the trained deep neural network that maps data having a second label and the first context to data having the second label and the second context, wherein the second context is different from the first context and the second label is different from the first label.