175 results about "Cross entropy" patented technology

The invention provides a multi-modal sentiment analysis method based on knowledge distillation and a dynamic fusion mechanism. The multi-modal sentiment analysis method comprises the following steps: pre-training a single-modal teacher model; a single-mode teacher model is used for guiding the learning of a multi-mode student model, and through an interactive knowledge distillation mechanism, the middle layer probability distribution of the teacher model is used as a target to learn the correlation between modes and the cross-mode characteristics; interactive knowledge distillation comprises three loss functions: firstly, calculating the difference of output distribution of output layers of a teacher model and a student model, and defining the difference as cross-modal knowledge distillation loss; secondly, adding alignment loss based on a real label, and constraining a prediction result of the student model to be close to a real emotion label in a cross entropy form; and finally, introducing label smoothing loss to soften the real label. According to the method, the pre-trained single-mode teacher model with relatively good performance is stored and is used for guiding the learning of the multi-mode student model, meanwhile, the loss function is introduced to optimize the multi-mode student model, so that the multi-mode student model is gradually aligned with the output distribution of the teacher model, and meanwhile, the adaptive capacity of the multi-mode student model to the modal heterogeneity is enhanced. The single-mode teacher model greatly reduces the complexity of the model, reduces the calculation amount, and has better performance in the field of multi-mode sentiment analysis.

The invention discloses a wire drawing machine operation state evaluation method and system based on deep learning, and relates to the technical field of intelligent manufacturing, and the method comprises the steps: deploying a multi-source sensor at a preset part of a wire drawing machine, and forming a multi-dimensional operation data flow; the deep learning evaluation model is trained based on the standardized time series data set, model parameters are optimized through a cross entropy loss function, the model learns the characteristic difference between a normal working condition and an abnormal working condition, and the deep learning evaluation model is trained based on the standardized time series data set. And inputting a multi-dimensional operation data flow collected in real time into the trained deep learning evaluation model to generate a dynamic evaluation report. According to the wire drawing machine operation state evaluation method and system based on deep learning provided by the invention, a visual operation state score can be output, detailed fault type probability distribution is given, and intelligent support is provided for equipment maintenance decision.

The invention discloses a hybrid power flow calculation method and device based on a cross entropy algorithm, and belongs to the technical field of optimization algorithm and machine learning crossing. The method comprises the steps that system modeling and parameter initialization are carried out, and a state space, an action space and a target function are defined according to a specific application scene; setting and initializing row parameters; entropy algorithm-guided sample generation and elite screening: iteratively optimizing power grid parameters through probability distribution, and quickly approaching a global optimal solution; the strategy optimization target of the strategy gradient algorithm is that a dynamic strategy network is constructed based on optimized power grid parameters, and the running state of the power grid is adjusted in real time to deal with uncertainty; performing sample multiplexing and distribution fusion of importance sampling: setting a double-buffer mechanism and performing sample fusion and multiplexing; and parameter updating and convergence judgment. According to the method, global search is carried out through the cross entropy algorithm, local optimization is carried out through the strategy gradient algorithm, the sample utilization rate is improved through the importance sampling method, and cooperative processing of global optimization and local optimization is achieved.

The invention discloses a cue word compression method and system based on TF-IDF and cross entropy, belongs to the technical field of large model cue word compression, and aims to solve the problems that redundant information is introduced into long cue words, the model efficiency is reduced and the cost is increased. To-be-compressed content is divided into sentences at the sentence level and then converted into embedded vectors, and the Euclidean distance is calculated in combination with problem vectors so as to screen related sentences; calculating a TF-IDF value at the word level through a word frequency and an inverse document frequency to extract keywords and recombine sentences; and selecting a reference model and a basic model at the Token level, identifying the key Token based on a cross entropy loss difference value, and splicing the key Token in sequence to generate a compressed cue word. According to the method, a complex calculation structure is avoided, the inference efficiency is improved while the semantic integrity is maintained, and the resource consumption is reduced.

The invention discloses a DeepLabV < 3 + >-based magnetotelluric two-dimensional inversion method, and belongs to the field of geophysical inversion methods, and the method comprises the steps: constructing a plurality of underground resistivity distribution models, and obtaining corresponding apparent resistivity and phase data through forward modeling calculation; designing a deep learning network based on a DeepLabV < 3 + > architecture, and enhancing the capability of capturing multi-scale geological features by using a cavity convolution and improved cavity spatial pyramid pooling (ASPP) module of the deep learning network; di ce Loss is adopted to replace a traditional cross entropy loss function, and the method is more suitable for the characteristics of anomaly and background information imbalance in geophysical observation data; according to the method, the network is trained and verified through the training sample set, and the optimal network parameters are obtained; the trained network is utilized to directly map the apparent resistivity and phase data of a test set into underground resistivity distribution, the end-to-end inversion process is realized, and the multi-scale feature extraction and semantic segmentation capabilities of the DeepLabV3 + network are utilized, so that compared with a traditional inversion method, the method has the advantages of being independent of an initial model, capable of obtaining a global optimal solution and the like.

The invention relates to an interpretable model robust training method based on topological regular terms, and belongs to the field of artificial intelligence safety. The method comprises the following steps: firstly, performing semantic preserving disturbance on an original sample to obtain a disturbance sample; secondly, inputting the original sample and the disturbance sample into a target model, and respectively calculating gradient values to generate corresponding interpretation images; secondly, calculating gradient differences before and after sample disturbance based on a cosine distance and an Euclidean distance, and extracting topological features of interpretation images before and after sample disturbance by using a persistent coherence method to quantify topological differences; and finally, taking the gradient difference and the topology difference as regular terms, forming total loss by the regular terms and the cross entropy loss, and dynamically adjusting the weight of the regular terms according to the proportion of each difference in the total loss. Aiming at the problems that the anti-interference performance of an existing method is influenced by only utilizing a gradient difference feature training model and a fixed regular term weight is difficult to adapt to generalization of a multi-type disturbance reduction model, the invention proposes that the robustness of model explanation is effectively improved by utilizing topological features of an explaining image.

A label noise estimation method based on a manifold regularization transfer matrix provided by the present invention pre-trains a first network in a second network, and after distilling a data set, inputs the obtained sub-data set into the second network to obtain the probability of the class to which the data instances in the sub-data set belong and obtain a transfer matrix related to the data instances; further calculates the cross-entropy loss of the second network according to the data instance labels, and combines an association matrix expressing the consistency of the data instances belonging to the same manifold and a penalty matrix of the data instances belonging to different manifolds to calculate the loss function of the second network; adjusts the loss function to reduce the training of the second network to obtain a trained second network, thereby completing the estimation of the class to which the data instances belong. The present invention can reduce the estimation error without affecting the approximation error of the transfer matrix, and experiments prove that the present invention can achieve excellent performance in label noise learning.

The invention discloses a crystal oscillator circuit fault classification technology based on a multi-source information fusion network, and belongs to the technical field of analog circuit fault diagnosis. Firstly, a multi-source data set of different measurement points of the crystal oscillator circuit is acquired; carrying out conversion from a time domain to a frequency domain on the data set by utilizing fast Fourier transform; extracting data fault features by using a convolutional neural network, and performing a trust distribution function under each piece of source data by using a softmax classifier; fusing different trust distribution functions under the multi-source data by adopting a D-S evidence theory to obtain a final diagnosis result and diagnosis probability output, and calculating cross entropy loss; and finally, training model parameters through back propagation to obtain a final diagnosis model. According to the method, the time-frequency transformation algorithm, the deep neural network algorithm and the information fusion algorithm are combined, a multi-source information fusion network is constructed, the defects of an existing diagnosis model in crystal oscillator circuit fault classification are overcome, and the accuracy and stability of crystal oscillator circuit fault diagnosis are remarkably improved.

The invention discloses an informatable meat segmentation method and device based on regularization and multi-model fusion. The method comprises the following steps: preprocessing data; credible polyp segmentation models based on PVTv2 and CNNs as encoders are constructed respectively; a loss function formed by combining cross entropy loss, KL divergence loss, Dice loss and newly defined evidence regularization is adopted for the two constructed information meat segmentation frameworks; an AdamW optimizer is adopted to train the two models at the same time; obtaining the final belief quality through a credible fusion rule; and Dice Score and mIoU are used as indexes for measuring the segmentation accuracy. According to the method, an evidence regularization item is introduced to improve the learning ability of the model on a zero evidence training sample. In addition, a credible fusion strategy is used, probability distribution and uncertainty output by multiple models are comprehensively considered, and the polyp segmentation accuracy is further improved.

The invention discloses a bearing fault diagnosis method based on a causal decoupling network, and relates to the technical field of computers. The method comprises the following steps: extracting overall characteristics of a sample bearing vibration signal; extracting cross-domain consistent causal factors of the overall features, and carrying out bearing fault classification on the causal factors; an extractor extracts non-causal factors of the overall features, performs fault classification on the non-causal factors, and performs domain classification on the non-causal factors; and determining the loss of the causal decoupling model based on the cross entropy loss of the causal branch and the non-causal branch, the cross-branch decorrelation loss, the maximum entropy loss and the branch specialization loss, and adjusting the parameters of the causal decoupling model until the loss of the causal decoupling model is minimum. The false correlation between the non-causal factors and the fault classification labels is blocked, and a trained causal decoupling model is obtained; and carrying out fault diagnosis on the to-be-classified bearing vibration signals through the trained causal decoupling model. The method can improve the diagnosis precision of the bearing fault.

The invention discloses a method for endowing a large language model with implicit thinking, and the method comprises the steps: introducing a potential token encoder which is isomorphic initialized with a basic model for a training data sample with an explicit reasoning chain of the large language model, collecting key information in the explicit reasoning chain through a special attention mask, generating a potential token, and transmitting the potential token to the basic model; limiting the potential tokens in a word list column space by using vocabulary probability superposition; discarding the potential token encoder, training a basic large language model (LLM) to autonomously generate a potential token, and adopting a combined target of KL divergence and cross entropy to stably optimize the target, so that potential reasoning collapses into an explicit answer at the reasoning tail end; and inputting a given question for the large language model LLM, generating a potential sequence in the LLM, and collapsing into a final answer to realize external hiding of intermediate reasoning content. According to the method, the efficiency, the precision, the stability and the engineering feasibility of potential reasoning are comprehensively improved.

The invention discloses a face-changing video detection method and system based on illumination feature decoupling. The method comprises the following steps of: framing a video and extracting an RGB (Red, Green and Blue) image of a detection area; sending the RGB image into a learnable local gravity mode extraction module, and extracting a local gravity mode image; the RGB image and the local gravity mode image are sent to a double-branch high-semantic feature extraction module, and illumination related features and illumination invariant features are extracted respectively; the illumination related features are sent to a contrast learning module based on relighting; sending the classification features into a classifier to obtain prediction probability distribution, and carrying out dichotomy supervision by using cross entropy loss; training a model and storing the model; and the model test loading model outputs a detection result of the to-be-detected video. According to the method, the illumination features are decoupled by using the domain generalization strategy, interference of domain-related illumination information is effectively suppressed, the characterization capability of the detection features is improved, and the method has a relatively good detection effect and generalization capability.

The invention relates to the technical field of power transmission engineering, and discloses an unbalanced sample galloping prediction method and system based on meteorological data, and the method comprises the steps: collecting the historical monitoring data of each monitoring point, and synchronously collecting the meteorological information of the corresponding monitoring point; monitoring devices and dates are taken as basic units, one piece of sample data is selected from daily data of each device, non-galloping samples in the sample data are grouped according to line numbers, and only samples corresponding to different spans are reserved in each group; according to line span characteristics, galloping high-incidence time periods and typical weather combinations, differential loss weights are given to sample data, and a weighted cross entropy loss function is constructed; and constructing a daily scale galloping prediction model, and training the model by using a weighted cross entropy loss function to realize daily scale prediction of a galloping event. By constructing a scientific sample extraction strategy, introducing a category weight mechanism and adopting a deep learning classification model, the recognition capability of the model on minority class events is improved.

The invention relates to the technical field of large language model privacy protection, in particular to a large language model black box forgetting method based on a double-auxiliary model, which comprises the following steps of: constructing the double-auxiliary model and generating output difference characteristics, analyzing parameter criticality through a Fisher information matrix to carry out self-matching fine tuning, calculating a weighted forgetting relevance score, and carrying out self-matching fine tuning on the weighted forgetting relevance score. And a forgetting response is generated based on score matching target model output. According to the method, the forgetting demand is judged through the output difference of the double-auxiliary model, the self-matching fine-tuning optimization parameter updating strategy is utilized, the forgetting relevance is calculated in combination with the weighted cross entropy and the target model loss, the target information is accurately forgotten in the black box environment, meanwhile, the performance of the model on the non-forgetting task is maintained, and the forgetting efficiency is improved. And the practicability and efficiency of privacy protection of the large language model are improved.

The invention relates to the technical field of computer vision, in particular to an adaptation method and system for continuous testing for differences between construction scene domains. According to the method, the similarity between Gram matrixes between adjacent domains is calculated, an elastic adjustment factor is set, the elastic adjustment factor is utilized, different weights are given to strong data enhancement and weak data enhancement, an elastic data enhancement strategy is provided, and an enhanced target domain image data set is input into a teacher model; updating the pseudo-tag by combining the elasticity regulation factor to obtain an elastic pseudo-tag; and inputting the target domain image data set into the student model to obtain a prediction result, constructing a global elastic symmetric cross entropy loss function based on the elastic adjustment factor, the cross entropy loss and the reverse cross entropy loss, updating student model parameters through the loss function, updating teacher model parameters, and finally obtaining a target model. According to the method, the construction scene monitoring model can adapt to complex domain changes when continuously learning test data, and the prediction result precision of the model in different environments is improved.

The invention provides a streaming three-dimensional semantic occupancy prediction method based on a Gaussian world model, and the method comprises the steps: employing explicit three-dimensional Gaussian representation, and explicitly modeling three decomposition factors of scene evolution in a three-dimensional Gaussian space; through a unified evolution layer module, modeling evolution of historical Gaussian and perception of newly complemented Gaussian at the same time, interacting Gaussian representation and visual input by adopting a three-dimensional convolution operation and a deformable attention mechanism respectively, and adding time sequence characteristic attributes to model historical information of Gaussian representation; and finally, predicting scene evolution by using all optimized Gaussian representations, performing three-dimensional semantic occupancy prediction of the current scene, and constraining the model through cross entropy loss and lovasz loss of a three-dimensional semantic occupancy prediction task. Compared with a traditional time sequence fusion method, the method has the advantages that scene evolution is explicitly modeled in the three-dimensional Gaussian space, the static consistency of scene evolution and dynamic object movement can be modeled, and the efficiency and effectiveness of model three-dimensional occupancy prediction are improved.

The invention provides a large voice model training method and device, equipment and a medium, and the method comprises the steps: inputting a training voice data subset into a large voice model, and obtaining the prediction probability distribution of the training voice data subset outputted by a large language model module in the large voice model; according to the prediction probability distribution and the real probability distribution, an entropy weighted cross entropy loss function is determined, and the entropy weight of the entropy weighted cross entropy loss function is determined based on the distribution entropy of the prediction probability distribution at the current moment; and with the purpose of minimizing an entropy weighted cross entropy loss function, updating parameters of the large language model module and training a voice recognition tag of the voice data subset. In the training method of the large voice model, entropy weight is added in an entropy weighted cross entropy loss function, the problem of uncertainty of prediction probability distribution is solved to a certain extent, iteration is performed by using a self-feedback signal of distribution entropy based on prediction probability distribution, and the voice recognition effect of the large voice model is continuously improved.

The invention relates to the technical field of intelligent fault prediction, and provides a multi-dimensional time sequence anomaly detection method for the process industry, which aims at historical and real-time multi-source data of the process industry, performs adaptive feature extraction to obtain better features through a feature selection module fusing mRMR and a self-attention mechanism, and improves the detection accuracy. Multivariable time series data double-fusion prediction calculation is completed through a sequence prediction model, the incidence relation of data at different time points and the incidence relation of different features at the same time are mined, and after cross entropy loss and optimizer training optimization are carried out, abnormal state judgment is finally achieved in combination with a dynamic threshold value. The invention further discloses a system used for the method, the method and the system can mine mechanism information implied by equipment from historical multi-source data in the process industry, time and features are subjected to relevance fusion respectively, and therefore the method and the system can be more suitable for the unsteady state and strong coupling conditions of the process industry; and particularly, a remarkable effect is achieved in a slag grinding system.

The invention discloses a temperature normalization knowledge distillation method and system based on multi-scale decoupling, and the method comprises the following steps: S1, inputting an image into a teacher model and a student model at the same time, respectively outputting logit feature maps by the teacher model and the student model, decomposing the global logit output of the teacher model into a plurality of local logit outputs, each local logit output corresponding to a specific region of the input image, extracting local logit information of the region from the feature map of the teacher model; the student model takes the local logit output by the teacher model as a learning result, and the learning result corresponds to the local logit of the teacher model on the same scale and the same area; s2, inputting the image into a pre-trained teacher model and a pre-trained student model respectively, calculating a standard deviation, performing local logit calculation on the teacher model and the student model according to an adaptive temperature mechanism to obtain probability distribution of the teacher model and probability distribution of the student model respectively, introducing a temperature scaling factor, and controlling the softening degree; and S3, combining the teacher and student model probability distribution with the normalized distillation loss, the semantic decoupling distillation loss and the cross entropy loss to calculate the total loss.

The invention discloses a control quantity-response quantity-based pseudo-visual anomaly detection method, which comprises the following steps of: determining a target measurement point capable of responding to a service scene, and analyzing and searching a control quantity related to the target measurement point; collecting and cleaning monitoring data containing a control quantity and a response quantity, and assembling the monitoring data into a data structure suitable for model training; processing the data by using an improved 1D U-Net network architecture, and carrying out model training on a supervised learning task with a fault sample by using a cross entropy loss function; for a learning task of an unsupervised sample, performing model training by using a reconstruction error; self-adaptive anomaly judgment and dynamic anomaly density labeling are carried out, finally, a detection model is trained and evaluated, and detection and early warning are carried out; the problems of low precision and high false alarm rate caused by difficulty in setting an early warning threshold value when a regression algorithm is used for prediction in the prior art are solved, and fault and normal classification judgment is realized through the classification model.

The invention relates to a time sequence enhanced and multi-granularity intention guided adversarial generation recommendation method, which comprises the following steps of: firstly, obtaining an original user commodity interaction sequence and interaction time data, and obtaining a time sequence enhanced sequence through a plurality of data enhancement modes; then, by constructing a plurality of cross entropy loss functions and comparison loss functions, model training is constrained from different angles, and user behavior patterns and potential correlation are fully mined; and finally, performing joint training on the feature coding module, and performing fine adjustment on the prediction module to obtain a recommendation model. The method focuses on solving the problems of data sparseness and insufficient model generalization ability of sequence recommendation (SR) under information overload. Under the background that information technology development causes serious information overload, although the SR is concerned, the SR faces many challenges; a traditional method depends on project prediction task optimization parameters, is easily influenced by data sparsity, and is difficult to capture real intentions of users and correlation between sequences; the method effectively improves the accuracy and reliability of a recommendation system, provides more accurate recommendation services for users, and has significant application value in the field of sequence recommendation.

The invention discloses a geological disaster hidden danger point distribution prediction method based on neural network integration. The geological disaster hidden danger point distribution prediction method comprises the following steps of S1, acquiring geological environment data and performing preprocessing; s2, constructing and training a heterogeneous model integrated by a graph convolutional neural network and a multi-layer perceptron; s3, combining residual connection and an attention mechanism to enhance the model feature recognition capability, and optimizing model parameters through a cross entropy loss function; s4, evaluating the heterogeneous model by adopting cross validation, calculating precision, mean square error and confidence coefficient, and dynamically distributing integrated weight; s5, weighting the output of the fusion model according to the integrated weight, and introducing Bayesian uncertainty estimation to generate a risk distribution probability graph; s6, quantifying the risk levels of the hidden danger points, mapping the hidden danger points into a two-dimensional spatial distribution map and visualizing the two-dimensional spatial distribution map And S7, screening hidden danger points of which the risk levels are higher than a threshold value, and outputting a prediction result. The invention provides a high-precision and reliable geological disaster hidden danger point prediction method through integrating a neural network and a deep fusion technology.

The invention provides a bearing fault diagnosis method and system, a storage medium and electronic equipment. The method comprises the following steps: acquiring source domain data and target domain data; denoising the source domain data and the target domain data; extracting features of the source domain data and the target domain data to obtain an initial feature vector; performing initialization processing on the initial feature vector; iteratively updating nodes of the source domain data and the target domain data to obtain a source domain graph and a target domain graph, and calculating a difference value; calculating the similarity loss of the two types, calculating the real distribution probability and the predicted distribution probability of the bearing type, obtaining a cross entropy loss function, obtaining a maximum mean value difference function, and calculating a total loss function; carrying out gradient calculation on parameters of the metric function, the source domain graph, the target domain graph, the nodes and the graph kernel model; and supervising the result, and performing joint optimization on the graph kernel model in combination with a graph kernel loss function. According to the method, the graph morphological characterization difference between the source domain graph and the target domain graph is reduced, and the generalization ability of the recognition model is improved.

The present invention discloses a test-time adaptation method for a calibration-free brain-computer interface, the method comprising the following steps: obtaining labeled training data from multiple source domain users to establish classification models for multiple categories; establishing the classification model, performing Euclidean alignment on the training data of each user, then merging all source domains, training multiple classification models using traditional cross-entropy loss, and obtaining multiple trained source domain models; obtaining the next test sample on the target domain data stream of the current user, performing incremental Euclidean alignment, and inputting the data into the trained multiple source domain models to obtain predicted probability values based on the multiple source domain models on the target domain data; using a spectral meta-learner method to perform integrated prediction of multiple models to obtain the predicted value of the current test sample. Based on the predicted probability value, each model is optimized separately by minimizing conditional entropy and using adaptive marginal distribution constraints within the batch to obtain optimized models adapted to the target domain; the present invention takes into account the real-time application of the cross-user brain-computer interface system, and adaptively adjusts the model without adding a calibration link.

The present invention realizes a commodity classification method based on distance geometry. By S1, the commodities and the relationships between the commodities are modeled into a graph, S2 obtains an initial embedding matrix, S3 learns the distance information between the commodities, S4 sets hyperparameters and the number of propagation times, S5 reduces the dimension of the matrix obtained after propagation to obtain a prediction matrix, S6 uses a cross-entropy loss function to calculate the loss and optimizes it with a backpropagation algorithm, S7 repeats until the classification accuracy of the algorithm on the validation set no longer improves within a certain number of iteration steps, then stops updating the model parameters; inputs the features of the test set commodities into the model, and finally obtains the output prediction. Thus, the technical effects of improving the classification accuracy, coping with the scenarios of homogeneous and heterogeneous matching diagrams, having interpretability in the airspace, and having advantages in terms of computational complexity are achieved.

The invention discloses an equipment deployment method based on precoding cross entropy optimization in an Internet of Things system, which comprises the following steps of: (1) defining a physical boundary and an initial parameter of an equipment deployment area, and initializing an equipment deployment space coordinate; (2) discretizing a continuous space into binary codes through a space pre-coding module and a position coding mapper, and compressing search dimensions; (3) generating candidate deployment position samples according to the probability distribution; (4) calculating the system performance of the candidate position by using a complex black box system and an evaluation function; (5) updating probability distribution based on an elite sample, and accelerating convergence to an optimal solution; and (6) judging an optimal deployment position. Compared with a traditional gradient optimization and linear enumeration method, the closed-loop global optimization of the equipment deployment position is realized through space coordinate discretization coding, probability-driven candidate sample generation, system evaluation and a cross entropy probability iteration updating mechanism; and the global optimality guarantee and the calculation efficiency are synchronously improved.

The invention discloses an open domain associated knowledge theme extraction method, and belongs to the technical field of machine learning, and the method comprises the steps: obtaining embedded information containing words and word contexts in sentences; inputting the embedded information containing the word itself and the word context into a pre-established iterative grid labeling model, and obtaining an iterative word label after a plurality of rounds of iteration; according to the obtained iteration word labels, calculating cross entropy loss between the prediction labels and the gold labels in each iteration round, and adding all the obtained cross entropy loss to obtain total cross entropy loss; designing constraint conditions corresponding to different requirements and task characteristics extracted from the associated knowledge theme, and calculating loss of the constraint conditions; and combining the total cross entropy loss and the constraint condition loss to obtain total loss, and outputting an open domain association knowledge topic extraction result according to the total loss. According to the method, the labeling cost can be reduced, and the accuracy of the associated knowledge topic extraction result is effectively improved.

The application discloses a kind of multi-modal large model learning method, system and storage medium based on instantaneous detection and rebalancing, method includes: S1, according to input sample construction multi-modal large model;S2, extract each single mode feature from training sample and calculate its prediction probability;S3, fusion all single mode features and obtain initial multi-modal feature and calculate its prediction probability;S4, compare the prediction probability difference of single mode and initial multi-modal, calculate rebalancing feature fusion weight;S5, based on rebalancing feature fusion weight calculation balanced multi-modal feature and calculate balanced multi-modal prediction probability;S6, calculate cross-entropy loss and update model parameter;S7, adjust the initial weight of next round feature fusion, return step S2 until all samples have participated in model training;The application can perceive and dynamically adjust the fusion weight between modes in time, to solve the mode imbalance problem in model training, to improve the accuracy and robustness of multi-modal large model.

The application relates to the technical field of semi-supervised learning, and discloses a method for semi-supervised learning by using adaptive double thresholds, which comprises the following steps: S1, for the labeled data, the weakly augmented data is input into a model, and a cross-entropy loss is obtained by taking the predicted result and the corresponding label; S2, when the model is trained by using unlabeled data, an adaptive threshold is extracted for each class, the fixed threshold and the class adaptive threshold are combined to form an adaptive double threshold, in addition to the high-confidence unlabeled data confirmed by using the fixed threshold, there is also the unlabeled data whose predicted value is less than the fixed threshold but greater than the extracted class adaptive threshold, different learning strategies are designed for the two types of unlabeled data; and S3, a new similar loss is proposed to further mine the information between similar unlabeled data, and the effective information in the unlabeled data is fully utilized.

This invention discloses an accelerated safety assessment method and system for low-altitude navigation systems based on alternating cross-entropy and importance sampling, relating to the fields of computer simulation and safety assessment technology. The method includes: constructing a probabilistic risk assessment model and initializing a proposal distribution; entering an alternating iterative loop: first, performing importance sampling assessment, calculating accident rate estimates and statistical errors; if the error does not meet accuracy requirements, initiating cross-entropy optimization, updating the proposal distribution parameters using simulation data; the new distribution is fed back to the next round of assessment. This loop continues until the results converge, and an assessment report is output. This invention, through a closed-loop feedback mechanism of alternating assessment and optimization, enables the sampling distribution to adaptively approximate the optimal distribution, intelligently focusing simulation resources on high-risk scenarios, solving the computational efficiency bottleneck of assessing extremely low probability events, achieving an order-of-magnitude improvement in assessment efficiency, and possessing high adaptability and engineering practical value.