19results about How to "Avoid negative transfer" patented technology

The invention provides a myocardial infarction positioning method based on cross-modal knowledge fusion. The method comprises the following steps: acquiring a target 1D-VCG signal and constructing a 2D-VCG image data set; extracting spatial features, time sequence features and pseudo tags from the 2D-VCG signals and the 1D-VCG signals through a teacher model and a student model; aligning and transforming the spatial features and the time sequence features through a cross-modal generator; and cross-modal knowledge fusion and MI positioning are realized through a cross-modal generator. According to the method, through knowledge enhancement and student enhancement strategies, the accuracy of myocardial infarction positioning is improved, and the overall framework of cross-modal learning is optimized through the adaptive weight method; the method is realized in a knowledge distillation form, spatial features of 2D-VCG are successfully fused into 1D-VCG time sequence analysis, and the positioning accuracy of myocardial infarction is remarkably improved by combining cooperative work of a teacher model, a student model and a cross-modal generator.

The invention belongs to the technical field of artificial intelligence, and particularly relates to a mixed identifier generation type recommendation method and system based on local collaborative context, and the method comprises the steps: obtaining a user historical interaction sequence, and carrying out the multi-granularity clustering of users, and obtaining a group of each user; constructing a static identifier based on the article content, constructing a dynamic identifier in combination with the user group information and the local interaction sequence, and fusing to generate a mixed identifier of each article; constructing an instruction fine tuning task, embedding group information into an instruction prompt word, and learning by using a large language model to generate a mixed identifier of a next article from a historical sequence; and generating a recommendation result based on the optimized large language model according to the historical sequence of the target user and the group information thereof. According to the method, the local collaborative context is fused through multi-granularity group estimation, and the mixed article representation is constructed in combination with static and dynamic identifiers, so that the problems of single user interest modeling and semantic deficiency of article representation are effectively solved, and the accuracy and personalized level of generative recommendation are improved.

The application discloses a kind of deep learning landslide identification method and system of fusing geology and mining prior knowledge, it is related to geological disaster remote sensing identification and deep learning technical field, the method includes: constructing the geological and mining prior knowledge graph of target coal mine area, and according to prior knowledge graph generates prior feature map;Acquire the remote sensing image to be identified, prior feature map is registered with remote sensing image in space, the center point of landslide candidate area in the remote sensing image is extracted using lightweight network, and landslide candidate area is obtained by adaptive cropping;The segmentation model after target domain meta migration learning training is input into landslide candidate area, and the landslide probability graph of each candidate area is obtained;Through channel attention mechanism, the landslide probability graph under multiple prior assumptions is adaptively fused, and the final landslide identification mask is generated, and the landslide disaster identification result is output according to landslide identification mask.In this way, the recognition accuracy and generalization ability are improved under the condition of sample scarcity.

The present invention discloses an early weak fault diagnosis method based on self-adjusting joint distribution adaptation, and relates to the field of mechanical engineering technology. The method comprises the following steps: step 1: collecting one-dimensional vibration signals of the source domain and early weak fault signals of the target domain. Step 2: building a Gabor-CNN backbone network. Step 3: re-dividing the two-dimensional image samples of the source domain after preprocessing. Step 4: statistically analyzing the distribution of target domain samples with the same pseudo-label, and calculating the average value to use as the threshold of the target domain samples for classification. Step 5: calculating the distribution average value corresponding to the target domain sample data before each round of backbone network classification, and calculating the ratio with the threshold in step 4, and using the ratio as the coefficient of the conditional distribution loss and the marginal distribution loss in the total loss function; and thus performing successive iterations to achieve the final classification. This method can realize the effective classification of early weak faults in the degradation stage of mechanical equipment, and prevent the high losses caused by serious faults in the later stage.

The present application belongs to the technical field of new energy lithium battery management, and particularly relates to a lithium battery life prediction method based on decomposition migration multi-dimensional Gaussian process, comprising: obtaining capacity observation data of historical batteries and in-service batteries, and constructing a capacity attenuation curve; inputting the curve into a decomposition migration multi-dimensional Gaussian process model, which decomposes the in-service battery capacity attenuation curve into a main trend part, a local fluctuation part and observation noise, migrates the main trend part of the in-service battery by sharing the corresponding latent Gaussian process in the main trend part of the historical battery capacity attenuation curve, and fits the local fluctuation part by using a convolution process; and probabilistically predicting the future capacity of the in-service battery based on the decomposition result. The present application selectively migrates the main trend, isolates the local fluctuation, avoids negative migration caused by capacity regeneration, improves prediction accuracy, reduces computational complexity by using a sparse covariance structure, and is suitable for cold start scenarios with sparse data.

Embodiments of the present specification provide a method and device for training a multi-task model, wherein the multi-task model comprises a backbone network for determining a user representation, and k head networks for performing k user prediction tasks based on the user representation. The method comprises: determining, based on m user samples, k sets of original gradient vectors of the k user prediction tasks with respect to the backbone network, wherein each user sample comprises a user feature and k user labels; mapping the k sets of original gradient vectors to a subspace of an original space in which the k sets of original gradient vectors are located, to obtain k sets of mapped gradient vectors; determining r weights corresponding to the r spatial dimensions of the subspace based on the component distribution of the k sets of mapped gradient vectors on the r spatial dimensions of the subspace, and performing weighted processing on the r dimensional components of each mapped gradient vector respectively by using the r weights to obtain k sets of weighted gradient vectors; and mapping the k sets of weighted gradient vectors back to the original space to obtain k sets of processed gradient vectors, which are used to update the network parameters of the backbone network.

The present application relates to the field of environmental monitoring and remote sensing information processing technology, and discloses a method and system for predicting the concentration of nutrient salts in a breeding area coupled with remote sensing data. The method comprises: constructing a physically constrained ecological hydrodynamic model to generate synthetic remote sensing-nutrient salt pairing data; fusing multi-source remote sensing and measured data to form a high-dimensional feature tensor; separating breeding sensitive features and environmental interference features through an attention-guided domain-invariant feature decoupling network; transferring prior knowledge of the source domain large-scale breeding area using a dynamic weight cross-domain knowledge distillation architecture; and finally outputting real-time high-precision nutrient salt concentration distribution through a lightweight regression head and time series smoothing. The system includes corresponding functional units. The present application improves the prediction accuracy and stability in small sample scenarios, suppresses negative transfer, and provides support for precision breeding and eutrophication early warning.

The embodiment of the specification provides a migration source selection method and system for short-term load migration prediction, wherein the method comprises the following steps: S1. Preprocessing migration source data according to the length of target source data to obtain a subsequence with the same length as the target source data; S2. Measuring the optimal transmission distance WD between the migration source and the target source by using the Wasserstein distance for each subsequence, and calculating the maximum information coefficient MIC between the migration source and the target source by using the maximum information coefficient method; S3. Constructing a WD-MIC curve with WD as the abscissa and MIC as the ordinate; and S4. Selecting the migration source with the maximum similarity as the final migration source by calculating the area under the WD-MIC curve as the similarity between the migration source and the target source. The application can effectively avoid negative migration in multi-source migration prediction.

This invention discloses a fault diagnosis method for rotating machinery based on hierarchical low-rank decoupling and manifold hybrid enhancement. This invention addresses the problem of poor generalization robustness of models under extreme variable operating conditions due to feature coupling and manifold holes. The method first acquires the original signal and converts it into a time-frequency map; then, it inputs it into a deep network containing hierarchical low-rank normalization units. The shallow layer randomly decouples and amplifies texture, while the deep layer utilizes gradient inversion for adversarial decoupling, actively stripping away specific operating condition styles. Next, manifold interpolation is performed on similar features at the bottleneck layer to fill manifold holes in a physically valid manner and maintain momentum prototypes. Finally, a prototype-based cosine classifier is constructed to immunize amplitude fluctuations, and a sharpness-aware algorithm combined with multiple metrics is used for two-stage smoothing optimization. This invention effectively decouples redundant operating condition features and overcomes minimum value traps, significantly improving zero-shot diagnostic accuracy in unknown variable operating condition scenarios.

This invention relates to a method, apparatus, device, and medium for cross-spectral stereo depth estimation. The method includes: inputting acquired left-eye thermal infrared and right-eye visible light images into a cross-spectral depth estimation network to obtain a disparity map. The cross-spectral depth estimation network includes: an encoder module for extracting features from the left-eye thermal infrared and right-eye visible light images respectively, obtaining thermal infrared features and visible light features; a guided fusion module for generating a pixel-level reliability weight map based on the right-eye visible light image, and fusing the visible light and thermal infrared features using the reliability weight map to obtain cross-modal fused features; and a decoder module for separating high and low frequencies in the cross-modal fused features, enhancing and stitching the obtained high-frequency and low-frequency features respectively, and obtaining a disparity map based on the stitched enhanced features. This invention can improve the robustness and accuracy of all-weather depth estimation.

The present invention relates to an adaptive document selection summary generation method based on transfer learning, and belongs to the field of artificial intelligence. In order to solve the problem of artificial intelligence-assisted archive research, the present invention uses a document selection method based on word distribution and a document selection method based on summary trend to filter a small amount of training documents of source documents and archives, screens out training data to train a first word vector generator, and obtains a basic summary generation model; the source documents and test documents are screened by an adaptive document selection method, and according to the characteristics of the test documents, a source document with strong correlation is selected to train a second word vector generator, optimizes the basic summary generation model, and constructs a customized summary generation model to more accurately meet the needs of specific archive categories. The present invention can realize automatic generation of summaries, promote archive intelligence, remove negative migration of archives, increase summary relevance, realize summary customization, and improve summary accuracy.

The application discloses a point cloud target detection model training method and device based on asymmetric feature consistency, relates to the technical field of automatic driving perception and computer vision, and comprises the following steps: acquiring source domain point cloud data, inputting a model to obtain a target detection loss, and respectively extracting geometric and intensity features; generating enhanced geometric features by disturbing the geometric features, and respectively mapping the geometric features, the enhanced geometric features and the intensity features into high-dimensional original geometric feature representation, enhanced geometric feature representation and intensity feature representation; determining geometric self-consistency loss based on the original geometric feature representation and the enhanced geometric feature representation; performing a stop gradient operation on the original geometric feature representation, and determining asymmetric feature alignment loss in combination with the intensity feature representation; and finally weighting and integrating the above losses into a total loss to update model parameters. By establishing one-way gradient blocking, the application not only utilizes intensity semantic auxiliary geometric feature learning, but also completely blocks noise pollution caused by cross-domain intensity deviation, and significantly improves the single-source domain generalization capability of the model.

The invention discloses a seismic phase identification method and system fusing meta learning and transfer learning, and belongs to the field of seismic phase identification, and the method comprises the following steps: training a neural network based on a large-scale seismic data set, and obtaining a pre-training model; on the basis of a pre-training model, performing meta-training on the model by adopting a plot-type sample comprising a global support set, a regional auxiliary support set and a query set, obtaining fusion loss by adopting double-flow knowledge injection and self-learning weighted fusion on the basis of the global support set and the regional auxiliary support set, updating a fast weight on the basis of the fusion loss, and obtaining a fast weight; calculating element loss based on the query set to update element parameters to obtain an element training model; migrating the meta-training model to a target area, and performing fine tuning based on the labeled waveform of the target area to obtain an adapted model; and inputting a seismic waveform to be identified into the adapted model, and outputting a P-wave and S-wave arrival time probability sequence.

The application discloses a kind of based on dynamic distribution self-adapting multi-source domain migration learning's remaining useful life prediction method, comprising the following steps: 1) given existing source domain and target domain degradation data;2) pre-processing is carried out to degradation data;3) the degradation characteristic representation of source domain and target domain degradation data is extracted;4) align the degradation characteristic distribution of each source domain and target domain, obtain the RUL label of multiple degradation characteristic representation of target domain;5) the RUL label obtained by each specific field predictor is fused as the final RUL prediction label, and the RUL label is fused target domain multiple degradation characteristics. Migration learning can utilize the similarity between data, tasks or models, apply the model and knowledge learned in the old field to the new field. The RUL prediction method based on migration learning trains the prediction model using the existing degradation dataset, and applies the learned knowledge to different working condition datasets to realize cross-domain RUL prediction.

The application belongs to the technical field of artificial intelligence, and specifically relates to a hybrid identifier generation type recommendation method and system based on local collaborative context, acquires a user historical interaction sequence, carries out multi-granularity clustering on the user to obtain a group of each user; a static identifier is constructed based on item content, a dynamic identifier is constructed in combination with user group information and local interaction sequence, and a hybrid identifier of each item is fused and generated; an instruction fine-tuning task is constructed, group information is embedded into an instruction prompt, a large language model is used to learn to generate a hybrid identifier of a next item from a historical sequence; and a recommendation result is generated based on the optimized large language model according to a historical sequence of a target user and group information of the target user. The application fuses local collaborative context through multi-granularity group estimation, combines static and dynamic identifiers to construct a hybrid item representation, effectively solves the problems of single user interest modeling and missing semantics of item representation, and improves the accuracy and personalized level of the generation type recommendation.

The invention relates to a transformable attention Transform micro-expression recognition method and system based on generative auxiliary domain adaptation, and the method comprises the steps: obtaining expression video sequence data, generating an optical flow tensor, and constructing a source domain data set and a target domain data set; constructing a residual error generator, mapping the source domain data set and generating an auxiliary domain data set; constructing a micro-expression recognition model, and performing training by using the source domain data set, the auxiliary domain data set and the target domain data set; meanwhile, constructing a composite loss function to obtain a trained micro-expression recognition model, and performing micro-expression recognition by using the trained micro-expression recognition model; through a generative auxiliary domain and three-domain collaborative alignment strategy, efficient migration of macro-expression knowledge to micro-expression tasks is realized, and the micro-expression recognition accuracy, the F1 score and the average recall rate are improved.

The invention discloses a personalized short message verification code pushing method based on multi-task learning, and the method comprises the following steps: collecting original user data of a user, and carrying out the preprocessing; inputting to an improved AdaTT network, and carrying out multi-modal feature fusion and weight distribution; calculating a risk probability value after executing dynamic gating fusion with the task proprietary features, and comparing the risk probability value with a preset threshold value; performing cross-task correction and selecting an optimal channel after dynamic gating fusion is performed on the channel task features; after dynamic gating fusion processing is executed, cross-task interaction is carried out, probability distribution is output, and an optimal time window is selected; performing joint judgment, constructing a joint loss function and updating an initial short message verification code pushing strategy; and executing the pushing operation of the short message verification code based on the personalized short message verification code pushing strategy. According to the method, the improved AdaTT network is adopted, personalized short message verification code intelligent pushing is achieved, and the method has the advantages of being high in safety, flexible in strategy and excellent in user experience.

The present invention is a resource allocation method based on transfer reinforcement learning, which is applicable to the hybrid multiple access scenario of uplink multi-cell, including: step 1: building a deep reinforcement learning network for resource allocation; step 2: performing MA-DRL interaction with the environment, in each round of interaction; step 3: performing the training process of MA-DRL, the training will be based on the experience blocks obtained by the interaction in step 2, the experience blocks obtained by each interaction with the environment are stored in the memory bank, and value labels are applied, and experience blocks with larger value labels are selected when extracting memory blocks; step 4: based on the resource allocation network, a transfer learning scheme based on MA-DRL is proposed. The resource allocation method proposed by the present invention aims to maximize the total rate of all users, can effectively improve the total rate of users, and can achieve a faster convergence speed after combining with transfer learning.