Abstract: This application relates to the artificial intelligence field, and discloses a neural network search method and a related apparatus. The neural network search method includes: constructing attention heads in transformer layers by sampling a plurality of candidate operators during model search, to construct a plurality of candidate neural networks, and comparing performance of the plurality of candidate neural networks to select a target neural network with higher performance. In this application, a transformer model is constructed with reference to model search, so that a new attention structure with better performance than an original self-attention mechanism can be generated, and effect in a wide range of downstream tasks is significantly improved.

Abstract: An antenna structure includes a first signal connector and a second signal connector. The antenna structure further includes a cavity antenna defined by a set of planar walls. The cavity antenna is coupled to the first signal connector and configured to emit a field polarized linearly in a first direction when driven by a signal at the first signal connector. The antenna structure further includes a dipole antenna defined by a pair of arms that are integrated with a wall of the cavity antenna. The dipole antenna is coupled to the second signal connector and configured to a field polarized linearly in a second direction offset from the first direction when driven by a signal at the second signal connector.

Abstract: The present application provides a precise ion implantation type modified geopolymer and an efficient preparation method thereof, wherein the precise ion implantation type modified geopolymer comprise: a geopolymer source material with a specific size, which implanted by Si and Al ions with a designated dose by an ion implanter. The geopolymer source material consists of a mixture of solid waste, alkali activator, and water.

Abstract: A method for preparing an ultra-fine and uniform acrylamide-based polymer hydrogel filament. The method comprises: formulating a reaction solution of acrylamide-based polymer under ice bath conditions, and drawing the reaction solution into a reaction tube by a syringe; parabolically bending the two ends of the reaction tube upward; hanging the reaction tube for 2 h-8 h, and performing a baking reaction to obtain an acrylamide-based polymer hydrogel; vacuum drying the reaction tube for 4 h-8 h to obtain a dry filament of acrylamide-based polymer; cutting the reaction tube to obtain a plurality of short polymerization tubes having a preset length; pushing the dry filament of acrylamide-based polymer out of the short tube by a stainless steel wire; purifying same in a soaking solution to obtain an expanded filament of acrylamide-based polymer; and drying same to obtain a acrylamide-based polymer hydrogel filament having a preset external diameter.

Abstract: Provided are catechol nanoparticles, catechol protein nanoparticles, and a preparation method and use thereof. The method includes: adding a tannin compound-containing natural herb medicine into water to obtain a mixture, and subjecting the mixture to heating reflux extraction to obtain a herb medicine extract and subjecting the herb medicine extract to fractionation to obtain the catechol nanoparticles.

Abstract: An apparatus and method in a wireless communication system and a computer readable storage medium. The apparatus comprises a processing circuit, which is configured to: configure a sounding reference signal (SRS) mapping structure for a user equipment on the basis of at least the number of antenna ports of the user equipment and the number of transmitting beams and/or receiving beams to be scanned; and notify the user equipment of the SRS mapping structure, wherein the SRS mapping structure at least comprises SRS resource settings for achieving uplink beam management and channel state information (CSI) acquisition at the same time, thereby reducing the system time delay and improving the uplink resource utilization rate.

Abstract: A method is provided for data processing performed by a processing system. The method comprises determining a set of first tokens for first data and a set of second token for second data, each token comprising information associated with a segment of the respective data, determining pair-wise similarities between the set of first tokens and the set of second tokens, each pair comprising a first token in the set of first tokens and a second token in the set of second tokens, determining, for each first token in the set of first tokens, a maximum similarity based on the determined pair-wise similarities between the respective first token and the second tokens in the set of second tokens, and determining a first similarity between the first data and the second data by aggregating the maximum similarities corresponding to the first tokens in the set of first set of tokens.

Abstract: The present disclosure relates to a radar ranging system based on a true random generator and a ranging method thereof The present disclosure adopts the following technical scheme: the system comprises a radar signal modulation and transmission unit and a radar signal receiving and processing unit, wherein the radar signal modulation and transmission unit inputs and stores a true random binary sequence generated by the true random generator in a memory, and cyclically outputs the true random binary sequence to a BPSK modulator for phase modulation. The receiving and amplifying unit receives the reflected radar signal. The down-sampling unit down samples the intermediate frequency signal generated by the mixers. The data processing unit converts the sampled analog signal into a digital signal and cross-correlates the signal with the true random binary sequence in the memory, so as to obtain the distance to the object side to be measured.

Abstract: This disclosure discloses lane line detection methods and devices. In an implementation, features extracted by different layers of the neural network are fused to obtain a fused second feature map, so that the second feature map obtained through fusion processing has a plurality of layers of features. The fused second feature map has a related feature of a low-layer receptive field and a related feature of a high-layer receptive field. Afterwards, an output predicted lane line set is divided into groups, where each predicted lane line in each group has an optimal prediction interval.

Abstract: A terahertz hollow core waveguide includes several successively cascaded waveguide units, and the waveguide units includes fiber core and cladding. The fiber core is composed of air, and the cladding is composed of dielectric rings, air rings, support strips, and an outer cladding. The medium rings and the air rings are successively surrounded on the outside of the fiber core, and the outer cladding is surrounded on the outside of the outermost air ring. All support strips in the same air ring of the same waveguide unit form a support strip group, and the support strips in the support strip group are arranged along the circumferential direction to connect two adjacent dielectric rings in the same waveguide unit or to connect the outermost dielectric ring and the outer cladding in the same waveguide unit.

Abstract: A neural network construction method and apparatus in the field of artificial intelligence, to accurately and efficiently construct a target neural network. The constructed target neural network has high output accuracy, may be further applied to different application scenarios, and has a strong generalization capability. The method includes: obtaining a start point network, where the start point network includes a plurality of serial subnets; performing at least one time of transformation on the start point network based on a preset first search space to obtain a serial network, where the first search space includes a range of parameters used for transforming the start point network; and if the serial network meets a preset condition, training the serial network by using a preset dataset to obtain a trained serial network; and if the trained serial network meets a termination condition, obtaining a target neural network based on the trained serial network.

Abstract: The disclosed embodiments concern methods, apparatus, systems and computer program products for determining and displaying pedigrees based on IBD data. Some implementations use a probabilistic relationship model to obtain various likelihoods of various potential relationships based on pairwise IBD data and pairwise age data. Some implementations build large pedigrees by combining smaller pedigrees. Some implementations display pedigree graphs with various features that are informative and easy to understand.

Abstract: A processing apparatus includes a collection module and a training module, the training module includes a backbone network and a region proposal network (RPN) layer, the backbone network is connected to the RPN layer, and the RPN layer includes a class activation map (CAM) unit. The collection module is configured to obtain an image, where the image includes an image with an instance-level label and an image with an image-level label. The backbone network is used to output a feature map of the image based on the image obtained by the collection module.

Abstract: A terahertz hollow core waveguide includes several successively cascaded waveguide units, and the waveguide units includes fiber core and cladding. The fiber core is composed of air, and the cladding is composed of dielectric rings, air rings, support strips, and an outer cladding. The medium rings and the air rings are successively surrounded on the outside of the fiber core, and the outer cladding is surrounded on the outside of the outermost air ring. All support strips in the same air ring of the same waveguide unit form a support strip group, and the support strips in the support strip group are arranged along the circumferential direction to connect two adjacent dielectric rings in the same waveguide unit or to connect the outermost dielectric ring and the outer cladding in the same waveguide unit.

Abstract: The present invention relates to a process of fabricating a multi-layer composite structure by 3D printing, said composite structure comprises at least one cured mortar layer formed by curing of a mortar composition, and at least one polyurethane layer formed by polymerization of a first polyurethane forming composition, wherein said mortar composition and said first polyurethane forming composition are dispensed individually and simultaneously via adjacent printing nozzles. The mortar composition is optionally modified by a second polyurethane forming composition.

Abstract: A neural network search method, apparatus, and device are provided, and relate to the field of artificial intelligence technologies, and specifically, to the field of automatic machine learning technologies. The method includes: A computing device obtains a dataset and N neural networks (S602), where N is a positive integer; and performs K evolutions on the N neural networks to obtain neural network obtained through the Kth evolution, where K is a positive integer (S604). In a process of each evolution, a network structure of a neural network obtained in previous evolution is mutated; and a candidate neural network is selected, based on a partially ordered hypothesis, from networks obtained through mutation.

Abstract: The present disclosure relates to a radar ranging system based on a true random generator and a ranging method thereof. The present disclosure adopts the following technical scheme: the system comprises a radar signal modulation and transmission unit and a radar signal receiving and processing unit, wherein the radar signal modulation and transmission unit inputs and stores a true random binary sequence generated by the true random generator in a memory, and cyclically outputs the true random binary sequence to a BPSK modulator for phase modulation. The receiving and amplifying unit receives the reflected radar signal. The down-sampling unit down samples the intermediate frequency signal generated by the mixers. The data processing unit converts the sampled analog signal into a digital signal and cross-correlates the signal with the true random binary sequence in the memory, so as to obtain the distance to the object side to be measured.

Abstract: Embodiments herein describe systems and methods to identify and treat pancreatic ductal adenocarcinoma (PDAC) based on cell of origin. Various embodiments comprise assessing transcriptomic data to identify gene expression profiles for the cancer. Based on type of cancer, additional embodiments treat individuals for PDAC. Further embodiments identify possible treatments for a PDAC tumor, by screening effective treatments, including drugs.

Abstract: This application provides a method and related apparatus for determining a neural network in the field of artificial intelligence. The method includes: obtaining a plurality of initial search spaces; determining M candidate neural networks based on the plurality of initial search spaces, where the candidate neural network includes a plurality of candidate subnetworks, the plurality of candidate subnetworks belong to the plurality of initial search spaces, and any two of the plurality of candidate subnetworks belong to different initial search spaces; evaluating the M candidate neural networks to obtain M evaluation results; and determining N candidate neural networks from the M candidate neural networks based on the M evaluation results, and determining N first target neural networks based on the N candidate neural networks. According to the method and the related apparatus provided in this application, a combined neural network with relatively high performance can be obtained.

Abstract: The disclosed embodiments concern methods, apparatus, systems and computer program products for determining and displaying pedigrees based on IBD data. Some implementations use a probabilistic relationship model to obtain various likelihoods of various potential relationships based on pairwise IBD data, and pairwise age data. Some implementations build large pedigrees by combining smaller pedigrees. Some implementations display pedigree graphs with various features that are informative and easy to understand.