Abstract: A continuous mining and delayed filling mining method for a deep ore body masonry structure is provided, comprising: dividing an ore body into ore blocks along a trend, internally dividing each ore block into stopes with square masonry structures, and reserving a rib pillar between the ore blocks; arranging an ore block conveyor belt gallery and a stope conveyor belt gallery at the lower parts of the ore blocks, arranging ore block crossheading and stope crossheading at the upper parts of the ore blocks, mining the stopes in the sequence from the foot wall to the hanging wall. In accordance with the present disclosure, adverse effects caused by deep high geo-stress and high geo-temperature on mining operation can be effectively overcome. The method has the advantages of low carbon and environmental protection, safety of recovery operation, high mechanization of stope operation, low labor intensity of manual operation and the like.

Abstract: A rock true triaxial dynamic compression-shear test equipment, and method under deep complex structure conditions, includes a counterforce framework, a pressure chamber, a base platform and four actuators. A pressure chamber transfer track is arranged on the base platform, and a pressure chamber transfer slipway is arranged on the tracks. The pressure chamber is a split structure. A pressure chamber base is located on the pressure chamber transfer slipway, and a pressure chamber barrel is provided with a pressure chamber barrel lift type bearing table and a pressure chamber barrel hoisting mechanism. The pressure chamber is provided with pressure chamber barrel and base packaging fixtures.

Abstract: A deep learning training method includes obtaining a training set, a first neural network, and a second neural network, where shortcut connections included in the first neural network are less than shortcut connections included in the second neural network; performing at least one time of iterative training on the first neural network based on the training set, to obtain a trained first neural network, where any iterative training includes: using a first output of at least one first intermediate layer in the first neural network as an input of at least one network layer in the second neural network, to obtain an output result of the at least one network layer; and updating the first neural network according to a first loss function.

Abstract: A neural network building method and apparatus are disclosed, and relate to the field of artificial intelligence. The method includes: initializing a search space and a plurality of building blocks, where the search space includes a plurality of operators, and the building block is a network structure obtained by connecting a plurality of nodes by using the operator; during training, in at least one training round, randomly discarding some operators, and updating the plurality of building blocks by using operators that are not discarded; and building a target neural network based on the plurality of updated building blocks. In the method, some operators are randomly discarded. This breaks association between operators, and overcomes a co-adaptation problem during training, to obtain a target neural network with better performance.

Abstract: Provided in the present invention are a compound as shown in formula I, various crystal forms thereof, a hydrate, a solvate or a pharmaceutically acceptable salt thereof, and a pharmaceutical composition. The compound of the present invention can effectively treat Alzheimer's disease or Down's syndrome, especially cognitive function impairment caused by Alzheimer's disease or Down's syndrome. Also disclosed in the present invention is use of a ubiquitin specific protease USP25 encoded and expressed by chromosome 21 for preventing, treating or ameliorating Alzheimer's disease or Down's syndrome.

Abstract: A neural network construction method and system in the field of artificial intelligence, to construct a target neural network by replacing a part of basic units in an initial backbone network with placeholder modules, so that different target neural networks can be constructed based on different scenarios. The method may include obtaining an initial backbone network and a candidate set, replacing at least one basic unit in the initial backbone network with at least one placeholder module to obtain a to-be-determined network, performing sampling based on the candidate set to obtain information about at least one sampling structure, and obtaining a network model based on the to-be-determined network and the information about the at least one sampling structure. The information about the at least one sampling structure may be used for determining a structure of the at least one placeholder module.

Abstract: This application relates to the field of artificial intelligence.

Abstract: A data processing method related to the field of artificial intelligence includes adding an architecture parameter to each feature interaction item in a first model, to obtain a second model, where the first model is a factorization machine (FM)-based model, and the architecture parameter represents importance of a corresponding feature interaction item; performing optimization on architecture parameters in the second model to obtain the optimized architecture parameters; and obtaining, based on the optimized architecture parameters and the first model or the second model, a third model through feature interaction item deletion.

Abstract: A motion behavior pattern classification method, system and device relating to the human motion recognition field. The method includes: S1, determining a candidate motion behavior pattern which includes a motion behavior pattern to be classified; S2, acquiring time series of behavior data of the candidate motion behavior pattern through a motion control sensor; S3, establishing an LSTM motion behavior pattern classification model through the time series; S4, predicting the motion behavior pattern to be classified through the LSTM motion pattern classification model, comparing the prediction result obtained using the sequence within the T time with the prediction result obtained using the sequence within the T+?t time using the iterative process to obtain the final prediction result, wherein T1?T?T2, T1>0, T2>T1, and the incremental step is set to be ?t, ?t>0. The technical solution of the present invention may improve accuracy while reducing the latency of the motion behavior pattern classification.

Abstract: This application provides a neural architecture search method, an image processing method and apparatus, and a storage medium. The method includes: determining a search space and a plurality of structuring elements, stacking the plurality of structuring elements to obtain an initial neural architecture at a first stage, and optimizing the initial neural architecture at the first stage to be convergent; and after an initial neural architecture optimized at the first stage is obtained, optimizing the initial neural architecture at a second stage to be convergent, to obtain optimized structuring elements, and building a target neural network based on the optimized structuring elements. Each edge of the initial neural architecture at the first stage and each edge of the initial neural architecture at the second stage correspond to a mixed operator including one type of operations and a mixed operator including a plurality of types of operations respectively.

Abstract: The present invention relates to a rock true triaxial dynamic compression-shear test equipment and method under deep complex structure conditions. The device comprises a counterforce framework, a pressure chamber, a base platform and four actuators. Pressure chamber transfer tracks are arranged on the base platform, and a pressure chamber transfer slipway is arranged on the tracks. The pressure chamber is a split structure. A pressure chamber base is located on the pressure chamber transfer slipway, and a pressure chamber barrel is provided with a pressure chamber barrel lift type bearing table and a pressure chamber barrel hoisting mechanism. The pressure chamber is provided with pressure chamber barrel and base packaging fixtures.

Abstract: An impeller includes a base plate, a hub protruding from the base plate, and a number of blades connected to the base plate and extending from a lateral surface of the hub toward an edge of the base plate. Opposite sides of each of the blades are a working surface and a non-working surface. The base plate is divided into a plurality of sub-plates by the plurality of blades. Each of the sub-plates is located between two adjacent blades and connected to the non-working surface of one of the two adjacent blades and the working surface of the other one of the two adjacent blades. A distance from an edge of each of the sub-plates to a center of rotation of the impeller is varied.

Abstract: The present invention discloses a network resource recommendation method, including: determining a first interest model according to a user behavior record in a current time window; calculating a similarity between any two interest models in the first interest model, a second interest model, and a third interest model, where a time window corresponding to the second interest model is adjacent to the current time window, the third interest model is determined based on all user behavior records in a reference time window set, the reference time window set includes N time windows prior to the time window corresponding to the second interest model; determining a fourth interest model and a recommendation algorithm according to the calculated similarity between any two interest models; generating a network resource recommendation list according to the fourth interest model and the recommendation algorithm; and making a recommendation according to the network resource recommendation list.

Abstract: A motion behavior pattern classification method, system and device relating to the human motion recognition field. The method includes: S1, determining a candidate motion behavior pattern which includes a motion behavior pattern to be classified; S2, acquiring time series of behavior data of the candidate motion behavior pattern through a motion control sensor; S3, establishing an LSTM motion behavior pattern classification model through the time series; S4, predicting the motion behavior pattern to be classified through the LSTM motion pattern classification model, comparing the prediction result obtained using the sequence within the T time with the prediction result obtained using the sequence within the T+?t time using the iterative process to obtain the final prediction result, wherein T1?T?T2, T1>0, T2>T1, and the incremental step is set to be ?t, ?t>0. The technical solution of the present invention may improve accuracy while reducing the latency of the motion behavior pattern classification.

Abstract: An impeller includes a base plate, a hub protruding from the base plate, and a number of blades connected to the base plate and extending from a lateral surface of the hub toward an edge of the base plate. Opposite sides of each of the blades are a working surface and a non-working surface. The base plate is divided into a plurality of sub-plates by the plurality of blades. Each of the sub-plates is located between two adjacent blades and connected to the non-working surface of one of the two adjacent blades and the working surface of the other one of the two adjacent blades. A distance from an edge of each of the sub-plates to a center of rotation of the impeller is varied.

Abstract: The present invention discloses a network resource recommendation method, including: determining a first interest model according to a user behavior record in a current time window; calculating a similarity between any two interest models in the first interest model, a second interest model, and a third interest model, where a time window corresponding to the second interest model is adjacent to the current time window, the third interest model is determined based on all user behavior records in a reference time window set, the reference time window set includes N time windows prior to the time window corresponding to the second interest model; determining a fourth interest model and a recommendation algorithm according to the calculated similarity between any two interest models; generating a network resource recommendation list according to the fourth interest model and the recommendation algorithm; and making a recommendation according to the network resource recommendation list.

Abstract: A method and a kit for the detection of Trichomonas infection by detecting N-acetyl-?-D-hexosaminidase activity in a sample, and a method and a kit for the detection of Trichomonas or Candida infection by detecting N-acetyl-?-D-hexosaminidase activity in a sample. The presence of a Trichomonas or a Candida infection can be distinguished by a pH of the sample.

Abstract: The invention relates to the new usage of puerarin in the field of pharmaceutical products, in other words, it relates to the application of puerarin in the preparation of drugs for P2X3 mediated pain/nervous system diseases. The experiment shows that puerarin can inhibit pain-related behavioral responses, subsequently immunohistochemical analysis, in situ hybridization, RT-PCR, protein blotting and other techniques were utilized to observe the inhibition of puerarin on mRNA and protein expression of P2X3 receptor in dorsal root ganglion of rats with neuropathic pain and in dorsal root ganglion and sensory nerve ending in the burn model of rats, it was found by using whole-cell patch clamp technique that puerarin can significantly reduce the electric current activated by P2X3 receptor agonist in the neuron of dorsal root ganglion of rats with neuropathic pain.

Abstract: The invention relates to a method and a kit for the detection of Trichomonas infection by detecting N-acetyl-?-D-hexosaminidase activity. The invention also relates to a method and a kit for selective detection of presence of Trichomonas or Candida infection.

Abstract: The invention relates to the new usage of puerarin in the field of pharmaceutical products, in other words, it relates to the application of puerarin in the preparation of drugs for P2X3 mediated pain/nervous system diseases. The experiment shows that puerarin can inhibit pain-related behavioral responses, subsequently immunohistochemical analysis, in situ hybridization, RT-PCR, protein blotting and other techniques were utilized to observe the inhibition of puerarin on mRNA and protein expression of P2X3 receptor in dorsal root ganglion of rats with neuropathic pain and in dorsal root ganglion and sensory nerve ending in the burn model of rats, it was found by using whole-cell patch clamp technique that puerarin can significantly reduce the electric current activated by P2X3 receptor agonist in the neuron of dorsal root ganglion of rats with neuropathic pain.