2327 results about "Global model" patented technology

The invention discloses an artificial intelligence machine vision image acquisition system, and the system comprises a multi-mode perception layer which integrates a self-adaptive optical module, inhibits metal reflection, captures a visible light to short wave infrared image, and captures a motion edge; the dynamic adaptive layer adopts an illumination compensation and motion compensation module to dynamically adjust camera parameters and micro displacement compensation, feeds back an illumination trend, outputs a motion vector to the cognitive layer, generates a confrontation sample through a GAN, simulates virtual defects in combination with a physical engine, and expands training data; the cognitive reasoning layer is used for deploying a dynamic routing network, distributing computing resources according to image complexity and optimizing feature extraction efficiency; reducing data deviation through anti-fact analysis, and generating a thermodynamic diagram to explain a detection basis; and the collaborative decision-making layer is used for rapidly screening samples by edge nodes, training a global model by cloud aggregated data, automatically triggering manual rechecking when the confidence coefficient of the model is insufficient, synchronously optimizing a training set and a causal reasoning module by a rechecking result, and improving the labeling efficiency through AR assistance.

The invention discloses a remote sensing image generation method based on a federal visual language model, which belongs to the technical field of machine learning and specifically comprises the following steps: receiving text instruction description by each client; extracting a multi-scale feature map from private remote sensing image data through a visual encoder, and generating a semantic embedding vector by text instruction description through a language encoder; inputting the semantic embedding vector and the multi-scale feature map into a dynamic attention mask generator to generate pixel-level space weight distribution; carrying out weighted fusion operation on the multi-scale feature map, and generating visual feature representation of text conditionalization; generating a remote sensing image according with the description of the text instruction through an image decoder; the client uploads model parameter increments of the visual encoder, the language encoder and the dynamic attention mask generator to the central server; the central server aggregates the model parameter increments, and distributes the updated global model parameters to each client; according to the method, the flexibility and semantic consistency of remote sensing image generation are effectively improved.

The invention discloses a power distribution room anomaly detection system based on cloud side-end cooperation, and belongs to the technical field of intelligent power grids. In order to solve the problems of high network bandwidth pressure, insufficient edge computing capability, low anomaly detection accuracy, difficulty in multi-source data fusion and the like caused by the adoption of an end-cloud direct connection architecture in an existing power distribution room monitoring system, the system comprises: a data acquisition layer configured with various heterogeneous sensors to acquire operating parameters and environmental data in real time; the edge storage and calculation layer carries out local real-time processing, anomaly detection, model training and visual display, an anomaly detection module of the edge storage and calculation layer carries out research and judgment on real-time data to generate early warning information, and a prediction and detection linkage module monitors an anomaly probability trend and adjusts a sampling frequency; the edge gateway realizes protocol conversion and data forwarding; and the cloud decision-making layer aggregates multi-edge node data, optimizes a global model through federal learning, and issues and updates a local model. The system is used for improving the accuracy, real-time performance and reliability of anomaly detection of the power distribution room, reducing the operation and maintenance cost and realizing intelligent operation and maintenance.

The invention provides a private weight self-adaptive heterogeneous data federated cooperative training method and system in the technical field of federated learning and privacy computing, and the method comprises the steps: S1, enabling each client to carry out the differential privacy operation on a local data set based on a private weight, and obtaining a desensitized data set, encoding the desensitized data set through a heterogeneous data encoding model; s2, performing semantic alignment on each coding vector through a contrast learning model to obtain an aligned vector set; s3, training a local model through the alignment vector set, generating a local gradient, extracting local model parameters, and uploading the privacy weight, the local gradient and local difference parameters to a server; and S4, the server trains the global model based on the local difference parameter and the global gradient, extracts the global model parameter and issues the global model parameter to each client for training. The method has the advantages that the compatibility, the flexibility and the efficiency of heterogeneous data federation cooperative training are greatly improved.

The invention discloses a federated learning-driven cross-domain supply chain elastic inventory optimization system and a method thereof, and aims at realizing inventory data collaboration among same-level enterprises or regional nodes through transverse federated learning and ensuring data security by adopting a self-adaptive differential privacy protection mechanism. The method comprises the steps of constructing a transverse federated learning network, locally performing data preprocessing, adding differential privacy noise, iteratively training a global model based on federated deep reinforcement learning, generating a transverse inventory allocation and replenishment decision, and performing model adaptive adjustment in real time based on key performance indicators. According to the method, multi-target balance is considered, inventory configuration is dynamically optimized through a multi-target reward function, the inventory turnover rate is remarkably increased, the inventory holding cost is reduced, the service level is improved, and the method is suitable for various scenes such as retail chain, manufacturing industry distributed storage and cross-regional logistics distribution; and global optimal inventory configuration is realized on the premise of ensuring data privacy.

The invention discloses a big data privacy protection modeling method and system based on federated learning and a block chain, and relates to the technical field of privacy protection and joint modeling. According to the method, homomorphic encryption, differential privacy, federated learning, secure multi-party computing and block chain technologies are fused, big data privacy protection and joint modeling are realized, encryption and dimensionality reduction are performed on original data through homomorphic encryption and differential privacy, an encrypted training sample of secure privacy is generated, a local model is trained on an encrypted data set through federated learning, and a big data privacy protection result is obtained. The method comprises the following steps: calculating aggregation parameters by using security multiple parties, constructing a verification network in combination with a block chain, ensuring credibility and integrity of model training, and finally, adding noise optimization performance for a global model by using differential privacy, testing generalization ability through cross validation, and determining a deployable privacy protection joint learning model, thereby breaking traditional data islands, promoting cross-mechanism data cooperation, and improving the privacy protection performance. Big data values are released, and data protection regulations and privacy requirements are met.

The invention provides a knowledge question and answer rapid processing method and system based on artificial intelligence, and relates to the field of artificial intelligence. A multi-modal knowledge graph is constructed, collected multi-source teaching data is fused through a mixed retrieval strategy, and the mixed retrieval strategy comprises semantic retrieval, vector retrieval and metadata retrieval; multi-level question and answer processing is executed based on an RAG enhancement framework, a multi-modal input intention is analyzed, cross-library joint retrieval is performed, and an optimization answer is generated in combination with a teaching scene; distilling the global model to a lightweight TinyBERT architecture, dynamically optimizing question and answer quality through a cognitive reinforcement learning framework, positioning a key document from a comprehensive retrieval list, evaluating an optimized answer, and reconstructing an answer with a key document verification score; according to the invention, the professional skill level of teachers and students in the fields of artificial intelligence and large model application can be improved, and the personalized requirements of teachers and students in teaching, scientific research and innovation courses can be met.

The invention discloses a multi-source equipment health collaborative operation and maintenance method in a complex environment based on edge federation, and relates to the technical field of equipment collaborative operation and maintenance. Vibration acoustic emission current waveforms are mapped to a unified time-frequency grid at an edge gateway, and an encrypted sparse index is generated and uploaded; training a global model by combining a graph regular base network with a gradient direction and a distribution distance, and injecting fault information by using a new working condition protection door; after being issued by a sparse adaptation layer of double-temperature-zone distillation and random projection compression, fine adjustment is carried out on site under few samples through temperature gradual fusion and reversible orthogonal mapping, and only unit gradient direction and health labels are uploaded; the center adopts entropy constraint Bayesian filtering to fuse information to generate a health index, a maintenance schedule and a spare part plan are formed by integer programming according to confidence intensity mapping risk popularity, a result is differentially pushed and audited, and a closed loop of collection, learning, evaluation and decision is realized.

The invention discloses a personalized federated learning optimization method and system, and mainly solves the problem of poor performance of an existing personalized federated learning model. The method comprises the following steps: establishing a communication link between a client and a server; each client receives a current global sharing model parameter broadcasted by the server, loads the current global sharing model parameter to a local model, and introduces a total loss function of a dynamic alignment strength definition model; training and optimizing local model parameters, and updating shared parameters by using the parameters; the client side calculates a self-adaptive aggregation weight based on the parameter updating quantity norm, the data volume weight and the synchronous frequency weight of the client side, and uploads the self-adaptive aggregation weight and the updated shared parameters to the server; and the server receives the parameter and weight information uploaded by the server, executes global model aggregation to obtain an updated global model, and outputs the global model reaching accuracy convergence or a preset training round on the verification set. According to the method, local personalization and global consistency can be balanced, the robustness and efficiency of global aggregation are improved, and the method can be used for processing scenes of high data isomerism and dynamic change of client participation states.

The invention provides a cross-park enterprise data collaborative analysis method based on federated learning, and relates to the technical field of distributed machine learning and data security, and the method comprises the steps that a central server distributes an initial global model and configuration parameters to each park node; the nodes execute local data feature alignment to generate standardized feature vectors; calculating dynamic collaborative factors of local data and global distribution; adjusting a training strategy based on the collaborative factors and updating model parameters; collecting model updating through an encrypted channel, and screening effective updating by adopting a dynamic aggregation offset threshold value; performing weighted aggregation to generate a new global model; and terminating the process when the cross-park convergence condition is met or the maximum round is reached. According to the method, heterogeneous data differences are eliminated through a dynamic feature alignment mechanism, dual-channel collaborative evaluation and adaptive security protection are combined, multi-park collaborative modeling efficiency and robustness are remarkably improved on the premise of guaranteeing data sovereignty, and the problems of feature space splitting, weak attack protection and node contribution imbalance are solved.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as financial science and technology and medical health, and discloses a federal collaborative optimization method, device and equipment and a medium. Each participant terminal executes instruction tuning training and value alignment training based on the initial global model, generates model parameters and uploads the model parameters, and the server side aggregates the received model parameters, generates an iterative global model and repeats the training process until a preset performance index is reached. According to the method, instruction tuning and value alignment training are carried out in parallel at the terminal side of the participant, and the optimization result is fused at the server side, so that the adaptability and training efficiency of the model to multi-source data are improved, the problems that in the prior art, the optimization target is single, and aggregation strategies are lack of collaboration are solved, and the practicability and generalization ability of the model are enhanced.

A federated learning system (100) for data protection in data sharing and collaboration, consisting of: a module for data acquisition and local preprocessing that is configured to clean, normalize and standardize local data sets at each participating node without transferring raw data externally; a local model training module configured to train a machine learning model on the pre-processed local dataset; a secure model update and encryption module configured to encrypt and secure model parameters or updates before transmission using privacy protection techniques; a federated aggregation and coordination module configured to aggregate encrypted updates from multiple participating nodes into a global model; a module for monitoring and ensuring data protection compliance, configured to enforce data protection budgets and audit protocols and to ensure compliance with data protection regulations; a performance optimization and resource management module configured to optimize communication, computation, and resource utilization across all nodes; and a module for global model delivery and feedback, configured to redistribute the aggregated global model to participants and integrate performance feedback for iterative improvements.

The invention relates to the technical field of intelligent regulation and control of a heat supply system, in particular to a heat supply system hydraulic unbalance real-time correction method and system based on federated learning, and the method comprises the steps: constructing a temperature prediction model based on an LSTM neural network, deploying federated nodes at each user side, collecting multi-dimensional heat supply data in real time, and carrying out the preprocessing; local model parameters are encrypted and uploaded to a central server through a federated learning framework, global model parameters are generated based on verification set error dynamic weighted aggregation, and nodes are reversely updated; the updated model is used for predicting the heat load in the next one hour, the valve opening degree is dynamically adjusted in combination with a hydraulic equilibrium algorithm, and whole-network flow and heat demand matching is achieved. According to the method, data privacy is guaranteed through federated learning and a block chain technology, edge calculation and LSTM time sequence prediction are combined, the problems that a traditional method is high in data transmission delay, high in privacy risk and poor in model generalization are solved, and the real-time response capability and the energy utilization efficiency of a heat supply system are remarkably improved.

The invention relates to a federated learning method, system and device for personalized differential privacy protection and a medium. The method comprises the steps that a central server initializes global model parameters and issues the global model parameters to clients; each client sets an initial value and an extreme value of a personalized privacy budget based on data characteristics of the client; the client performs local training, cuts the gradient in the training process, and adds corresponding Gaussian noise processing based on the current personalized privacy budget; the central server performs weighted aggregation on the model parameters uploaded by the clients to update a global model, and issues the updated global model parameters to the clients for a new round of local training; and the central server dynamically adjusts the personalized privacy budget of each client based on the reward factor, and then allocates the personalized privacy budget to each client for local training again until a global model meeting a preset requirement is obtained. The method can be widely applied to the field of distributed machine learning data security.

Disclosed in the present invention are a large language model federated fine-tuning method and apparatus based on gradient compression. The method comprises the following steps: constructing, on the basis of a gradient tensor generated during fine tuning of a large language model, a raw data set having a time series relationship, performing inference by means of an autoencoder to obtain a reconstructed gradient data set, and constructing a reconstruction loss function to optimize the autoencoder; and initializing a base model of the large language model as a global model at a server end, the server end updating the global model to a client, using a pre-trained encoder to obtain a compressed gradient at the client, and at the server end, using a pre-trained decoder to decode and aggregate the compressed gradient, and then updating the global model. The present invention can improve the fine-tuning efficiency of the large language model and reduce computing resource requirements while ensuring data privacy protection, and is suitable for application scenarios such as communication optimization improvement and privacy protection enhancement in the process of scientific computing-oriented large model fine-tuning and training.

The invention discloses an urban multi-mode digital twin data asset security management method, and relates to the technical field of urban digital twin and data security, and the method comprises the steps: firstly building a unified access gateway, converging multi-department and multi-type data, and generating sensitivity annotation and privacy level metadata; performing differential privacy or desensitization and dynamically calibrating parameters at the local part of the data provider according to the labels; bookkeeping access, authorization and training on a permission alliance chain, and performing fine-grained permission control; through on-chain authorization, each node only uploads disturbed and encrypted model update, and a global model is obtained after secure aggregation; the governance layer continuously monitors logs, budget and model indexes and triggers strategy closed-loop regulation and control; and multi-agent reinforcement learning of a privacy agent and a performance agent is further introduced, a unified multi-target optimizer is formed through distillation, privacy, federation and permission strategies are refreshed online, and collaborative analysis and compliance sharing operation without data out of a domain are realized.

An efficient heterogeneous federated learning method based on hybrid distillation includes: initializing, by a server, global model parameters, and setting a preset total number of training rounds and a number of clients participating in each of the training rounds; loading local datasets in the clients respectively, performing random transformations on the local datasets to generate client distillation data for the clients, sampling multiple sub-networks from an original network of each client, training each sub-network on the client distillation data to obtain updated local model parameters of each client, and uploading the updated local model parameters to the server; and receiving, by the server, the updated local model parameters, performing, by the server, server distillation based on the updated local model parameters and a preset auxiliary dataset to obtain updated global model parameters and an updated global model, and sending, by the server, the updated global model to the clients.

The invention relates to a landslide grading early warning method based on multi-modal data change characteristics. The method comprises the following steps: establishing a mountain digital twinborn body; the method comprises the following steps: deploying a multi-node sensor network in a target area, configuring an edge computing unit, collecting geological data in real time, and screening the geological data based on mountain digital twin to form effective local data; constructing a lightweight multi-modal neural network model at each node, dynamically searching hyper-parameters by using an ant colony optimization algorithm, and generating an encryption model weight update quantity packet; the central server dynamically calculates node weights according to disaster feature vectors output by the digital twins, generates a global model through weighted aggregation, and directionally distributes and updates the global model; real-time monitoring data and a model prediction result are fused, millimeter-level disaster evolution simulation is executed through a variable step size physical engine, an advanced early warning signal is triggered when a deduced prediction risk exceeds a threshold value, and the edge model adaptability, federal aggregation precision and early warning advancement are remarkably improved.

The invention relates to the technical field of unmanned aerial vehicle clusters, and provides a federal reinforcement learning task allocation method and system for a heterogeneous unmanned aerial vehicle cluster, and the method comprises the steps: collecting the stability performance parameters of unmanned aerial vehicles in real time, carrying out the normalization processing of the stability performance parameters based on a preset weight factor, and obtaining the resource capability indexes of the unmanned aerial vehicles; training a reinforcement learning model, calculating a weight according to the resource capability index, and performing weighted fusion on model parameters to generate a global model; dynamic change information in a task area is collected in real time through an unmanned aerial vehicle sensor, a task allocation scheme of the unmanned aerial vehicle is planned and executed through a multi-target optimization algorithm, and the task execution state of the unmanned aerial vehicle is obtained; and calculating an efficiency value of the task allocation scheme according to the task execution state and the environment data, and when the efficiency value is lower than a preset threshold value, dynamically optimizing the task allocation strategy to obtain an optimized task allocation strategy. The task execution efficiency and the resource utilization rate of the heterogeneous unmanned aerial vehicle cluster in a complex dynamic environment are improved.

The invention relates to the technical field of federated learning, in particular to a dynamic federated mutual learning method and system for balancing individuation and generalization, and the method specifically comprises the following steps: each client carries out the preprocessing of data to be processed of a model, and carries out the strong enhancement and weak enhancement processing; inputting the data subjected to strong enhancement processing into a shared model, inputting the data subjected to weak enhancement processing into a private model, and performing iterative training on the two models; related parameters of the shared model after each round of iterative training and a difference item between two model parameters are uploaded to a federation server; the federated server adopts a multi-dimensional adaptive aggregation strategy to obtain an updated global model, and returns the updated global model to each client to replace the shared model in the next round of training; and finally generating a generalization result and a personalized result. According to the method, the private-shared model architecture is constructed, and dynamic federated mutual learning is carried out in combination with the federated server, so that balance and collaborative improvement of individuation and generalization performance can be realized.

The invention provides an energy optimization scheduling method and device, a storage medium and computer equipment, and relates to the technical field of energy scheduling, during energy scheduling, original data of each edge node is collected, and after a data hash value and key metadata of the original data are determined, blockchain fragmentation storage is performed, so that the energy scheduling efficiency is improved. Meanwhile, original data are encrypted into edge data by adopting a quantum key distribution method, so that the data cannot be tampered; next, edge data training is adopted to obtain a digital twin sub-model corresponding to each edge node, then federal aggregation is performed to form a global model, decentralized cooperation is realized, and carbon energy flow is simulated in real time; and then, dynamically optimizing the global model to generate a scheduling instruction, if the scheduling instruction is not in an expected range, performing incremental optimization correction on the global model until the scheduling instruction generated by the global model is in the expected range, and issuing the scheduling instruction to each edge node to perform energy scheduling. Therefore, millisecond-level real-time scheduling response can be realized while data privacy security is ensured.

The embodiment of the invention discloses a multi-microgrid regulation and control method and device based on federal hierarchical reinforcement learning, and a medium, belongs to the technical field of smart grids, and solves the problem of low microgrid dispatching precision. Each microgrid uploads a model gradient parameter corresponding to the local scheduling model to a central server; the central server determines an aggregation weight according to the similarity between the micro-grid data distribution characteristics so as to update a global model; the central server dynamically divides each micro-grid into a plurality of micro-grid clusters according to model gradient similarity, and constructs a leader-follower game model in each micro-grid cluster; the central server outputs a first scheduling strategy corresponding to each micro-grid based on the updated global model, the plurality of micro-grid clusters and the leader-follower game model; and each micro-grid periodically performs secondary optimization on the first scheduling strategy based on the local feature data to obtain a second scheduling strategy corresponding to the current micro-grid.

The invention discloses an Internet of Vehicles federated learning excitation method and system based on a Nash game and a coalition game, and the method comprises the steps: initializing a vehicle client set in an Internet of Vehicles environment, and enabling a server to broadcast global model parameters to all clients; a two-stage Nash game model is established, the first stage is used for determining an alliance set in the vehicle client set, and the second stage is used for bargaining to optimize resource contribution; negotiating and determining an optimal data contribution ratio and an optimal resource contribution ratio based on the server side and the alliance set by using the two-stage Nash game model; and the server allocates rewards to the alliance set according to the optimal data contribution proportion and the optimal resource contribution proportion. The method can adapt to high dynamics of the Internet of Vehicles, and fair and efficient resource allocation is realized.

The invention discloses an energy storage battery health state evaluation method and system based on federated learning, and relates to the technical field of energy storage batteries. The method comprises the following steps: training a local health state evaluation model based on a federal loss function containing physical prior constraints at each client device, and enhancing sparse working condition data by adopting a local generation model to generate local model update; in the central server, performing value-guided heterogeneous aggregation processing, performing weighted aggregation on local model update submitted by the client, and generating a global health state assessment model; in the central server, executing digital twinborn consistency calibration, and performing post-aggregation fine tuning on the global model by using a reference signal generated by a cloud digital twinborn body; and communicating between the client equipment and the central server by adopting an event triggering and gradient sparse quantization compression mechanism, and deploying an adaptive differential privacy policy based on gradient inversion auditing. According to the method, the problems of high data privacy leakage risk, low model precision under heterogeneous data and high communication overhead in the prior art are solved.

A federated object detection learning method based on representation enhancement and weighted aggregation under cloud-edge-terminal environment comprises the steps of: 1) building a centralized federated learning framework under cloud-edge-terminal environment; 2) locally conducting representation enhancement training to strengthen model learning for few-shot category after receiving a model from the server at the client; 3) carrying out the weighted aggregation for client models in accordance with sample distribution to obtain the global model after receiving models from all clients at the server. With regard to the problem of existing federated object detection learning on low global model accuracy and weak generalization ability, the present invention can improve the accuracy and generalization ability of global object detection model.

The invention discloses a cross-domain privacy protection method and system for advertisement recommendation, equipment and a medium, and the method specifically comprises the steps: carrying out the encryption matching of user behavior data and anonymized equipment data, and generating an initial cross-domain joint feature vector; fusing the initial cross-domain joint feature vector and the disturbance feature vector to form a target cross-domain joint feature vector; splitting a pre-trained advertisement recommendation model into a feature coding sub-module and a reasoning sub-module, deploying the feature coding sub-module to a user adjacent edge node, and retaining the reasoning sub-module in a user local device; and based on the target cross-domain joint feature vector, performing calculation of the feature coding sub-module and calculation of the reasoning sub-module, and uploading the encrypted hidden layer feature vector to a federated learning aggregation server for global model updating. According to the method, cross-domain data utilization and user privacy protection in an advertisement recommendation process are realized, and effective feature vectors are generated for personalized advertisement recommendation while data are guaranteed not to be out of a domain.

The invention discloses a personalized federated learning method based on knowledge fusion distillation and a storage medium, and belongs to the technical field of personalized federated learning, and the method comprises the steps: a server carries out the pre-training and distributes a diffusion model to each client, and generates a local synthesis data set, initializing a global model as a student model of a client to perform knowledge fusion distillation training, and guiding the student model training by the teacher model optimized in the last round; and after training is completed, the client uploads student model parameters to the server for federal aggregation to update the global model, and meanwhile, the personalized feature extraction capability and reliability of the next round of teacher model are enhanced by using synthetic data, so that a closed-loop learning framework of global cooperation and local personalized collaborative optimization is formed. According to the method, the problems of client model drifting, performance attenuation and convergence rate slowing caused by data heterogeneity can be solved.

The invention discloses a medical data management method and system based on multi-modal fusion and privacy protection, and the method comprises the steps: receiving multi-modal medical data, and executing cross-modal embedding learning to generate a joint embedding vector; constructing a patient multi-modal graph based on the joint embedded vector, and detecting abnormal inconsistency between the cross-modal data by using a graph attention network to generate an abnormal detection result; a privacy protection strategy is dynamically adjusted according to the anomaly detection result and the access context information, and desensitization processing is conducted on the medical data; and constructing a global model through security aggregation based on the desensitized medical data to generate an optimal governance action and apply the optimal governance action to a data management process. Through organic cooperation of multi-modal fusion, anomaly detection, adaptive privacy protection, security aggregation and dynamic strategy optimization, a closed-loop medical data governance platform is successfully constructed, and the problems of multi-modal data fragmentation, privacy compliance conflict and dynamic governance deficiency are effectively solved.

The invention discloses a federated learning method based on quantum transfer learning, relates to the cross technical field of quantum computing and federated learning, and aims to solve the problem of performance bottleneck of traditional federated learning under non-independent identically distributed data. The local model of each client is fused with a classic convolutional layer and a quantum convolutional layer; during local training, firstly, classic features are extracted by a classic convolutional layer, and then the classic features are coded and input into a quantum convolutional layer to generate quantum features; and the two features are spliced and then classified, and model parameters are uploaded. And the server side adopts a FedAvg algorithm to aggregate parameters so as to update the global model. According to the method, a transfer learning mechanism is introduced to optimize the initialization of the quantum layer, and an experimental result shows that the classification performance and robustness of the method are remarkably superior to those of a traditional federated CNN model on a non-IID image data set, and the method is particularly suitable for a privacy protection cooperative computing scene.

The invention relates to the technical field of intelligent dialogues, and discloses a dialogue marketing strategy optimization method based on artificial intelligence, and the method comprises the following steps: collecting the voice, text, facial expression and physiological signals of a user in real time through a multi-modal perception assembly of a terminal device, constructing a dynamic emotion map, and extracting a multi-modal feature vector; a cross-modal information processing module is utilized to align the multi-modal data through a comparative learning algorithm, and causal relationship description of user behaviors and strategies and strategy risk scores are generated; cooperatively training a global strategy model through differential privacy and homomorphic encryption technologies; combining the dynamic emotion map and a causal model to generate an emotion adaptive dialogue script, and optimizing strategy selection through a reinforcement learning algorithm; and dynamically updating the emotion map, the risk score and the global model through a closed-loop feedback mechanism to form a real-time optimized strategy generation system. According to the invention, the practicability of artificial intelligence to real-time services can be improved.