Abstract: A multi-task neural network system is described. The system includes a shared neural network configured to receive as input a text span from a clinical note, and for each of one or more text segments in the text span, processing the text segment to generate a set of text segment embeddings. The system further includes a segmentation neural network configured to, for each of the one or more text segments, process the respective set of text segment embeddings to determine whether the text segment is a section title or not. The system further includes a section type classification neural network configured to, for each of the one or more text segments, process the respective set of text segment embeddings to classify the text segment into a section type of a plurality of section types.

Abstract: According to one aspect, a method includes obtaining a sensor assembly that includes at least a first camera and a second camera, and positioning the sensor assembly in an enclosure of a sensor testing system. The enclosure has a first enclosure target and a second enclosure target affixed thereon, and includes a sensor arrangement. The sensor testing assembly further includes a computing arrangement and a data acquisition arrangement, The method also includes performing a first test on the sensor assembly by providing commands to the sensor assembly using the computing arrangement, and monitoring the sensor assembly during the first test. Monitoring the sensor assembly includes obtaining data from the sensor assembly and/or the sensor arrangement, and providing the data to the data acquisition arrangement. Finally, the method includes processing the data, wherein processing the data includes determining whether the data indicates that the sensor assembly passes the first test.

Abstract: This application discloses a bidirectional vent valve, a battery, and an electrical device. The bidirectional vent valve includes: a valve seat, where the valve seat is configured to form a gas duct, and the gas duct includes a first end and a second end; a bidirectional valve assembly, where the bidirectional valve assembly is movably disposed in the gas duct, and is configured to open or close the gas duct through movement relative to the gas duct; and a first elastic component, configured to exert a force on the bidirectional valve assembly to close the air duct, and, when an air pressure difference between the first end and the second end of the gas duct is greater than a first threshold, the bidirectional valve assembly is configured to overcome the force of the first elastic component to open the gas duct.

Abstract: A method for determining a three-dimensional tortuosity of a loose and broken rock-soil mass, includes the following steps: a particle grading curve of the loose and broken rock-soil mass is obtained by utilizing a particle size analysis, and followed by calculating an equivalent particle size and an average particle size; a porosity of the loose and broken rock-soil mass is obtained by utilizing a moisture content test, a density test, and a specific gravity test; the three-dimensional tortuosity of the loose and broken rock-soil mass is obtained by utilizing the equivalent particle size, the average particle size and the porosity of the loose and broken rock-soil mass. The method has the advantages of simple logic, accuracy and reliability, simple and fast parameter determination, and has high practical value and promotion value in the field of environmental protection and ecological restoration technology.

Abstract: An LDPC encoding method and a system for error code detection. In the method and system, partial syndromes using a user portion and a low density parity check matrix are calculated, a parity portion of a codeword is calculated using the partial syndromes and using a quasi-cyclic matrix, the parity portion is generated by segment processing of the quasi-cyclic matrix, and the user portion and the parity portion are concatenated to complete the codeword.

Abstract: A display panel includes: a liquid crystal cell; an optical layer configured to transmit part of light incident onto the optical layer whose polarization direction is parallel to a transmission axis of the optical layer, and reflect a remaining part of the light incident onto the optical layer; a first polarization structure configured to transmit part of light incident onto the first polarization structure whose polarization direction is parallel to a transmission axis of the first polarization structure, and absorb a remaining part of the light incident onto the first polarization structure; and a second polarization structure configured to transmit part of light incident onto the second polarization structure whose polarization direction is parallel to a transmission axis of the second polarization structure, and absorb a remaining part of the light incident onto the second polarization structure.

Abstract: To improve error correction when errors occur in consecutive bits of user data, the user data can be stored in an interleaved manner. Data of a data unit can be interleaved to generate a permutated data unit. A checksum of the permutated data unit can then be calculated, and an error correction code can be generated over the data unit and the checksum. The error correction code can also be interleaved to generate a permutated error correction code. The permutated data unit, the checksum, and the permutated error correction code can then be concatenated to generate a storage data unit for storage in a memory.

Abstract: Devices, systems, and methods for reducing a latency of a decoder in a non-volatile memory are described. An example method includes receiving a noisy codeword that is based on a transmitted codeword generated from a low-density parity-check (LDPC) code, the LDPC code having an associated parity matrix comprising a plurality of columns of circulant matrices, performing a sorting operation that sorts the plurality of columns of circulant matrices in a descending order of a first quality metric to generate a plurality of sorted columns of circulant matrices, the first quality metric indicative of a number of errors in a corresponding column of circulant matrices, and iteratively processing the plurality of sorted columns of circulant matrices to determine a candidate version of the transmitted codeword.

Abstract: The present disclosure provides an apparatus and a method for hydrate-based wastewater treatment and cold storage. The apparatus for hydrate-based wastewater treatment and cold storage includes a water chilling unit, a hydrate-based cold storage tank, an airflow disturbance device, a water layer positioning system, a spraying system, a suction filtration system, a heat exchange system, and a system monitoring device. The water chilling unit provides a secondary refrigerant at a low temperature. The secondary refrigerant flows through an evaporator coil in the hydrate-based cold storage tank for heat exchange. The airflow disturbance device induces hydrate nucleation. The water layer positioning system positions a contact surface between a water layer and a hydrate former after hydrate decomposition to facilitate drawing and separation of the treated upper water layer. The spraying system and the suction filtration system enhance the solid-liquid separation efficiency to improve the removal rate of pollutants.

Abstract: A hybrid inference facility receives a sequence of data items. For each data item, the facility: forwards the data item to a server; subjects it to a local machine learning model to produce a local inference result for the data item; and the local inference result to a queue; aggregates the inference results contained by the queue to obtain an output inference result; and removes the oldest inference result from the queue. The facility receives from the server cloud inference results each obtained by applying a server machine learning model to one of the data items forwarded to the server. For each received cloud inference result, the facility substitutes the cloud inference result in the queue for the local inference result for the same data item.

Abstract: The present disclosure provides a method for processing defect information of a product, which includes the following steps of: acquiring defect information on a current film layer and defect information on historical film layers; determining whether defect information exists at a target location of the historical film layer if defect information exists at a target location of the current film layer; if defect information exists for a corresponding location to the target location in at least one of the historical film layers, deleting the defect information detected at the target location in the current film layer; and if no defect information exists for the target location in any of the historical film layers, retaining the defect information detected at the target location in the current film layer.

Abstract: This invention provides a system-oriented and fully-controlled connected automated vehicle highway system for various levels of connected and automated vehicles and highways. The system comprises one or more of: 1) a hierarchical traffic control network of Traffic Control Centers (TCC's), local traffic controller units (TCUs), 2) A RSU (Road Side Unit) network (with integrated functionalities of vehicle sensors, I2V communication to deliver control instructions), 3) OBU (On-Board Unit with sensor and V2I communication units) network embedded in connected and automated vehicles, and 4) wireless communication and security system with local and global connectivity. This system provides a safer, more reliable and more cost-effective solution by redistributing vehicle driving tasks to the hierarchical traffic control network and RSU network.

Abstract: A memory system or a method for estimating channel information to be used for error decoding. The memory system or the method a) performs one or more read operations on a page selected from among the plurality of pages using a target read threshold, b) obtains the target read threshold, a historical read threshold voltage set associated with failed read operations of the selected page, checksum values, and asymmetric ratios of ones count and zeros count which are associated with the historical read threshold voltage set, c) provides the obtained target read threshold, historical read threshold voltage set, checksum values and asymmetric ratios as input information to a neural network, and d) predicts, by the neural network, channel information at the target read threshold based on the input information and a set activation function.

Abstract: A scheme for determining a flipping energy used in a bit-flipping decoder. The flipping energy is determined based on: a weight of at least one check node coupled to a column; a syndrome as a product of a noisy codeword and a parity check matrix; and a hard decision value of a previous iteration and a channel output value associated with the column.

Abstract: The present disclosure discloses a time service and positioning network system for an underground chain-shaped large space in a coal mine. The system adopts a base station-level network internal time synchronization method which takes an optical fiber time service as a main part and takes a satellite time service as an auxiliary part to construct a time service and positioning network system with an unified time within the ad-hoc network base stations, based on the ad-hoc network base stations linearly arranged in the underground chain-shaped large space in the coal mine. Through the network construction, the network stabilization and the network application, the system unifies the time reference of the base station-level network and improves the positioning accuracy of the network application by utilization of the base station-level network internal time synchronization method.

Abstract: A method and a system for operating a deep neural network. In the method and system, a subset of floating-point values are used to represent weights in the DNN; the floating-point values are quantized onto a flexible-power-of-two (FPoT) alphabet; values in the FPoT alphabet are listed in a plurality of regions; and an empty region among the plurality of regions is merged to neighbour regions to output dusters of the weights in merged regions, the merged regions having respective centroids and boundary lines in between.

Abstract: Anti-peep display assembly and display device are provided. The anti-peep display assembly includes a backlight module, a first birefringence control type liquid crystal cell, a second birefringence control type liquid crystal cell, and a display liquid crystal cell. Along a thickness direction of the anti-peep display assembly, the first birefringence control type liquid crystal cell and the second birefringence control type liquid crystal cell are between the backlight module and the display liquid crystal cell, and the first birefringence control type liquid crystal cell is between the second birefringence control type liquid crystal cell and the backlight module. A first polarizer is arranged on a side of the display liquid crystal cell away from the second birefringence control type liquid crystal cell, and a second polarizer is arranged between the display liquid crystal cell and the second birefringence control type liquid crystal cell.

Abstract: A screen projection control method, storage medium and communication apparatus are provided. The screen projection control method includes: a screen projection receiving end receiving screen projection request information from a first screen projection sending end; determining, according to the screen projection request information, whether a third screen projection sending end that has triggered an interference-free mode of screen projection exists in at least one second screen projection sending end that is performing screen projection, wherein enabling the interference-free mode comprises stopping accepting a screen projection request; and if the third screen projection sending end that has enabled the screen projection interference-free mode exists, the screen projection receiving end sending response information for refusing the screen projection request to the first screen projection sending end.