Abstract: The disclosure provides a semi-supervised method and apparatus for public opinion text analysis. The semi-supervised method includes: first acquiring a public opinion data set, and preprocessing the data set; performing a data augmentation algorithm on preprocessed samples to generate data augmented samples; generating category labels for the unlabeled samples in the data set in an unsupervised extraction and clustering manner; calculating similarities of word vector latent semantic spaces and performing linear interpolation operation to generate, according to an operation result, similarity interpolation samples; constructing a final training sample set; adopting a semi-supervised method, inputting the final training sample set into a pre-trained language model to train the model to obtain a classification model; and predicting the test set by using the classification model to obtain a classification result.

Abstract: The present disclosure provides a neural network computing-oriented modeling method and apparatus for distributed data routing. The method includes the following steps: S1, designing the distributed attribute of a physical tensor: abstracting a mapping relationship between a logic tensor and the physical tensor into three distributed attributes including a broadcast attribute, a scatter attribute and a local reduction attribute; S2, deducing the distributed attribute of an output tensor: specifying the distributed attribute of an input tensor, and then deducing the legal distributed attribute of the output tensor according to the known distributed attribute of the input tensor; and S3, judging, according to the distributed attribute situation, whether an intermediate communication primitive needs to be inserted to obtain the distributed attribute of a local physical tensor.

Abstract: The disclosure discloses a method of executing dynamic graph for neural network computation and the apparatus thereof. The method of executing dynamic graph includes the following steps: S1: constructing and distributing an operator and a tensor; S2: deducing an operator executing process by an operator interpreter; S3: constructing an instruction of a virtual machine at runtime by the operator interpreter; S4: sending the instruction to the virtual machine at runtime by the operator interpreter; S5: scheduling the instruction by the virtual machine; and S6: releasing an executed instruction by the virtual machine. According to the method of executing dynamic graph for neural network computation and the apparatus thereof provided by the disclosure, runtime is abstracted to be the virtual machine, and the virtual machine acquires a sub-graph of each step constructed by a user in real time through the interpreter and schedules, the virtual machines issues, and executes each sub-graph.

Abstract: The present disclosure is directed to methods and systems for knowledge distillation.

Abstract: Disclosed are an intermediate representation method and apparatus for parallel execution of graph computation. The method includes the following steps: S1: compiling a neural network into a computational graph on a computer; S2: defining branch states of tensor variables in the computational graph; S3: defining a data dependency relationship of the tensor variables in the computational graph; S4: defining a control dependency relationship of the tensor variables in the computational graph; S5: building a data dependency relationship graph of the tensor variables in the computational graph; S6: building a control dependency relationship graph of the tensor variables in the computational graph; and S7: transforming control dependencies into data dependencies. The present application derives, based on the dependency relationship, a parallel computing method that can execute the branch threads in parallel in the global computational graph, and optimizes the compilation efficiency of the computational graph.

Abstract: The present disclosure discloses a method for execution of a computational graph in a neural network model and an apparatus thereof, including: creating task execution bodies on a native machine according to a physical computational graph compiled and generated by a deep learning framework, and designing a solution for allocating a plurality of idle memory blocks to each task execution body, so that the entire computational graph participates in deep learning training tasks of different batches of data in a pipelining and parallelizing manner.

Abstract: The disclosure discloses a method of neural network model computation-oriented intermediate representation and apparatus thereof. The method includes the following steps: S1, parsing an input model file so as to acquire topological structure information of a neural network; S2, constructing a logical computation graph; S21, inferring physical layout information of each operator in the logical computation graph; S22, inferring meta attributes of each operator in the logical computation graph; S23, inferring description information of input and output logical tensors of each operator in the logical computation graph; S3, constructing a physical computation graph; S31, generating a physical computation graph, etc.

Abstract: Disclosed is a method for distributed type training adaptation and apparatus in a deep learning framework and an AI accelerator card. The method includes the following steps: S1: the deep learning framework supports single-card configuration in a newly added AI accelerator card, and sub-steps thereof are as follows: S11: the deep learning framework supports new hardware; S12: the deep learning framework supports a device thread of the new hardware; S13: the deep learning framework supports a memory operation of the new hardware; and S14: the deep learning framework supports an operator kernel function of the new hardware; S2: the deep learning framework supports multi-card configuration in the newly added AI accelerator card; S3: the deep learning framework supports tensor segmentation and multi-card distribution; and S4: the deep learning framework supports multi-card collective communication in the newly added AI accelerator card.

Abstract: Disclosed are a pedestrian re-identification method and apparatus based on local feature attention. The method includes the following steps: S1: obtaining an original surveillance video image data set, and dividing the original surveillance video image data set into a training set and a test set in proportion; and S2: performing image enhancement on the original surveillance video image training set to obtain enhanced images, and converting the enhanced images into sequence data.

Abstract: Disclosed is a method for distributed type training adaptation and apparatus in a deep learning framework and an AI accelerator card. The method includes the following steps: S1: the deep learning framework supports single-card configuration in a newly added AI accelerator card, and sub-steps thereof are as follows: S11: the deep learning framework supports new hardware; S12: the deep learning framework supports a device thread of the new hardware; S13: the deep learning framework supports a memory operation of the new hardware; and S14: the deep learning framework supports an operator kernel function of the new hardware; S2: the deep learning framework supports multi-card configuration in the newly added AI accelerator card; S3: the deep learning framework supports tensor segmentation and multi-card distribution; and S4: the deep learning framework supports multi-card collective communication in the newly added AI accelerator card.

Abstract: Disclosed is a method for meta-knowledge fine-tuning and platform based on domain-invariant features. According to the method, highly transferable common knowledge, i.e., domain-invariant features, in different data sets of the same kind of tasks is learnt, the common domain features in different domains corresponding to different data sets of the same kind of tasks learnt in the network set are fine-tuned to be quickly adapted to any different domains. According to the present application, the parameter initialization ability and generalization ability of the universal language model of the same kind of tasks are improved, and finally a common compression framework of the universal language model of the same kind of downstream tasks is obtained through fine tuning. In the meta-knowledge fine-tuning network, a loss function of the domain-invariant features is designed in the present application, and domain-independent universal knowledge is learn.

Abstract: Disclosed are a labeling method and apparatus for named entity recognition of a legal instrument. The method includes steps: step S1: acquiring a legal text, and transforming the legal text into an index table; step S2: outputting a sentence feature encoding result; step S3: performing training and prediction; step S4: obtaining a set; step S5: obtaining a multi-head score transfer matrix; step S6: obtaining a score transfer matrix corresponding to the legal text; step S7: determining a recognized nested entity; and S8: constructing an entity labeling template by using the recognized nested entity. According to the present disclosure, a user tries to complete recognition of nested entity labeling by changing an input of the BERT model, and a multi-head selection matrix labeling thought of the present disclosure is used to relieve the difficulty in recognizing a long text and a nested entity in an NER task to a larger extent.

Abstract: A system that includes: a force sensor assembly adapted to monitor a load as applied on a subject’s knee joint when the force sensor assembly remains in direct contact with the subject’s lower extremity and the load is monitored from inside a main magnet of an MRI scanner; a mobile unit comprising tracks configured to adjust a position of the force sensor assembly; a stationary base on which the mobile unit and the force sensor assembly are located, the mobile unit translatable solely axially on the stationary base; and a processor coupled to the force sensor assembly and programmed to read information encoding the load being monitored by the force sensor assembly, wherein an MRI scan of the knee j oint is initiated only when a predetermined load has been applied to the subject’s knee joint for a pre-determined period of time.

Abstract: Disclosed is a method for automatically compressing multi-task oriented pre-trained language model and a platform thereof. According to the method, a meta-network of a structure generator is designed, a knowledge distillation coding vector is constructed based on a knowledge distillation method of Transformer layer sampling, and a distillation structure model corresponding to a currently input coding vector is generated by using the structure generator; at the same time, a Bernoulli distribution sampling method is provided for training the structure generator; in each iteration, each encoder unit is transferred by Bernoulli distribution sampling to form a corresponding coding vector; by changing the coding vector input to the structure generator and a small batch of training data, the structure generator and the corresponding distillation structure are jointly trained, and a structure generator capable of generating weights for different distillation structures can be acquired.

Abstract: Disclosed are an automatic compression method and platform for a pre-trained language model based on multilevel knowledge distillation. The method includes the following steps: step 1, constructing multilevel knowledge distillation, and distilling a knowledge structure of a large model at three different levels: a self-attention unit, a hidden layer state and an embedded layer; step 2, training a knowledge distillation network of meta-learning to generate a general compression architecture of a plurality of pre-trained language models; and step 3, searching for an optimal compression structure based on an evolutionary algorithm.

Abstract: A system that includes: a force sensor assembly adapted to monitor a load as applied on a subject's knee joint when the force sensor assembly remains in direct contact with the subject's lower extremity and the load is monitored from inside a main magnet of an MRI scanner; a mobile unit comprising tracks configured to adjust a position of the force sensor assembly; a stationary base on which the mobile unit and the force sensor assembly are located, the mobile unit translatable solely axially on the stationary base; and a processor coupled to the force sensor assembly and programmed to read information encoding the load being monitored by the force sensor assembly, wherein an MRI scan of the knee joint is initiated only when a pre-determined load has been applied to the subject's knee joint for a pre-determined period of time.

Abstract: Disclosed is a method for meta-knowledge fine-tuning and platform based on domain-invariant features. According to the method, highly transferable common knowledge, i.e., domain-invariant features, in different data sets of the same kind of tasks is learnt, the common domain features in different domains corresponding to different data sets of the same kind of tasks learnt in the network set are fine-tuned to be quickly adapted to any different domains. According to the present application, the parameter initialization ability and generalization ability of the universal language model of the same kind of tasks are improved, and finally a common compression framework of the universal language model of the same kind of downstream tasks is obtained through fine tuning. In the meta-knowledge fine-tuning network, a loss function of the domain-invariant features is designed in the present application, and domain-independent universal knowledge is learn.

Abstract: Disclosed are an automatic compression method and platform for a pre-trained language model based on multilevel knowledge distillation. The method includes the following steps: step 1, constructing multilevel knowledge distillation, and distilling a knowledge structure of a large model at three different levels: a self-attention unit, a hidden layer state and an embedded layer; step 2, training a knowledge distillation network of meta-learning to generate a general compression architecture of a plurality of pre-trained language models; and step 3, searching for an optimal compression structure based on an evolutionary algorithm.

Abstract: Disclosed is a method for automatically compressing multi-task oriented pre-trained language model and a platform thereof. According to the method, a meta-network of a structure generator is designed, a knowledge distillation coding vector is constructed based on a knowledge distillation method of Transformer layer sampling, and a distillation structure model corresponding to a currently input coding vector is generated by using the structure generator; at the same time, a Bernoulli distribution sampling method is provided for training the structure generator; in each iteration, each encoder unit is transferred by Bernoulli distribution sampling to form a corresponding coding vector; by changing the coding vector input to the structure generator and a small batch of training data, the structure generator and the corresponding distillation structure are jointly trained, and a structure generator capable of generating weights for different distillation structures can be acquired.

Abstract: Disclosed is a meta-knowledge fine tuning method and platform for a multi-task language model. The method is to obtain highly transferable shared knowledge, that is, meta-knowledge, on different data sets of tasks of the same category, perform interrelation and mutual reinforcement on the learning processes of the tasks of the same category that correspond to different data sets and are in different domains, so as to improve the fine tuning effect of downstream tasks of the same category on data sets of different domains in the application of the language model, and improve the parameter initialization ability and the generalization ability of a general language model for the tasks of the same category.