1892 results about "Differential privacy" patented technology

The invention relates to the technical field of judicial data security, and discloses a judicial system confidential data security circulation method based on a block chain technology. Collecting multi-source judicial data, identifying sensitive information through a large language model, and performing differential privacy desensitization processing; constructing an SM4 encryption and TLS 1.3 end-to-end secure channel, generating a data hash fingerprint, and writing the data hash fingerprint into an alliance chain evidence based on a PBFT consensus mechanism; designing a multi-modal classification engine to perform feature extraction and intelligent classification; establishing a hybrid authority model to integrate an XACML strategy and a Kafka queue, and implementing dynamic authority control in combination with an RBAC / ABAC mechanism; a hierarchical encryption storage architecture is constructed, and homomorphic encryption retrieval and erasure code distributed storage are adopted; constructing a judicial knowledge graph based on a BERT model; and deploying a block chain auditing system, and combining LSTM anomaly detection and a DREAD risk assessment model to form a closed-loop risk control system. According to the method, the problems of security risk and privacy disclosure in judicial data cross-department circulation are solved.

The invention discloses a risk control credit monitoring method based on cloud computing, and belongs to the technical field of cloud computing, and the risk control credit monitoring method based on cloud computing comprises the following steps: S1, collecting user transaction data and behavior track data in real time; s2, cleaning and standardizing the data; s3, constructing a multi-dimensional risk assessment model based on machine learning; s4, dynamically generating a credit score according to the risk characteristics; s5, triggering an early warning mechanism for abnormal transactions in real time; and S6, generating a visual risk control report and updating a monitoring strategy. According to the method, multi-source heterogeneous data are integrated through federated learning, hierarchical privacy protection is realized in combination with homomorphic encryption and differential privacy, risk assessment real-time performance is improved by using a hybrid cloud resource scheduling and dynamic model updating technology, and a compliance audit closed loop is constructed based on a block chain and interpretability analysis.

The invention discloses a computer task scheduling method based on artificial intelligence, and the method comprises the following steps: 1, data collection: employing a double-flow feature fusion mechanism, and generating global feature representation containing long-term dependence and an instantaneous state; step 2, generating a global optimization scheduling strategy: constructing a hierarchical federal reinforcement learning system, dividing a cluster into a plurality of super nodes through an enhanced spectral clustering algorithm, independently training a Dueling DQN network by each super node, performing global strategy cooperation by adopting Shapley value weighted aggregation and differential privacy protection, and generating a scheduling strategy of global optimization; distilling a global strategy into a lightweight decision tree through a strategy distillation technology, and deploying the lightweight decision tree to a physical node; 3, task priority control and elastic resource allocation are carried out, wherein elastic control over resource allocation is carried out through a dynamic time slice bank mechanism; and 4, self-adaptive evolution: establishing a closed-loop optimization system, and carrying out strategy self-evolution by adopting a double-layer optimization architecture.

The invention discloses an artificial intelligence data privacy protection system based on a block chain and federated learning, and relates to the technical field of block chains and federated learning, and the system comprises a block chain module which employs a main chain-side chain double-chain architecture, a main chain stores a global model hash value and node reputation evaluation data, and a side chain module is used for storing node reputation evaluation data; the side chain stores the encrypted local model parameters through a fragmentation technology; the federated learning module comprises a dynamic difference privacy algorithm and a gradient ternary processing unit, and is used for adding noise to the gradient in a local training stage and converting the gradient into a ternary numerical format; the privacy protection module is used for integrating homomorphic encryption and zero-knowledge proof technologies and realizing ciphertext aggregation and verification of model parameters; and the malicious node detection module is used for identifying abnormal gradient update based on cosine similarity and Multi-Krum algorithm, and is linked with node reputation data in the block chain module. Through a system architecture and a privacy protection mechanism, the efficiency and performance of federal learning are improved while data privacy is ensured, and the method has a wide application prospect.

The invention discloses a human resource management method and system based on data security. The method comprises the steps that multi-source human resource data are collected through a standardized interface, and sensitive levels are marked in a classified mode; calling a dynamic encryption engine based on the sensitive level and the data type, performing asymmetric encryption on the structured data, and performing hybrid encryption on the unstructured data; a dynamic hierarchical access control strategy is generated in combination with user roles and service scenes, and the permission range is adjusted in real time; laws and regulations such as GDPR and CCPA are analyzed through a compliance rule base, data operation legality is automatically verified, and illegal behaviors are blocked; k-anonymity and differential privacy technologies are adopted to desensitize sensitive data, and privacy protection and data availability balance are ensured; and recording a full-process operation log based on the block chain and generating a non-tampering audit report. The system comprises a data classification acquisition module, a dynamic encryption engine module, an access control engine module and the like. According to the method, the problems of data protection rigidness, compliance response lag and privacy-utility imbalance of a traditional HR system are solved.

The invention relates to the technical field of data privacy protection, and discloses a cross-platform social privacy collaborative protection system based on federated learning and a block chain. The system comprises a federal learning initialization module, a private data encryption module, a cross-platform data synchronization module, a block chain consensus verification module and an intelligent contract execution module. Global model initialization parameters are generated through a multi-party security aggregation algorithm, user data privacy is protected through hierarchical encryption, data are synchronized through a Hash time lock protocol and an intelligent contract, model updating is verified through an improved Byzantine fault-tolerant algorithm, and a privacy protection rule is triggered based on a differential privacy noise injection algorithm. The system effectively solves the problem of cross-platform social privacy protection, guarantees data security and privacy, improves federal learning reliability, optimizes data sharing and utilization, and is suitable for various cross-platform social scenes.

The invention belongs to the field of intelligent edge systems and privacy enhancement computing, and particularly relates to a privacy protection-oriented robot large model cloud edge end collaborative reasoning and federated learning system, which comprises a cloud server layer used for deploying a large-scale pre-training model and executing complex reasoning and global federated learning coordination; the edge calculation layer is used for deploying an intermediate layer model and executing local data aggregation, privacy protection processing and intermediate feature calculation; the terminal equipment layer is used for deploying a lightweight model and executing data acquisition, primary processing and lightweight reasoning; the federated learning framework is used for optimizing the model; the privacy protection module is used for integrating data localization, differential privacy, homomorphic encryption, secure multi-party computing and a block chain verification mechanism; the adaptive allocation module is used for dynamically adjusting computing resources. According to the method, the problems of privacy leakage risk, computing resource limitation, network delay, insufficient data isolation and the like of the traditional AI service in a robot scene are solved, and efficient privacy protection and data security isolation are realized.

The invention discloses a real-time health risk prediction method and system based on a dynamic knowledge graph, and relates to the technical field of medical information. The method comprises the following steps: carrying out multi-modal fusion and privacy protection preprocessing on medical and nursing heterogeneous data, and realizing semantic consistency of cross-mechanism data based on an entity alignment method of a cross-modal graph neural network; based on a hierarchical federated learning framework, local model parameters are subjected to hierarchical encryption and aggregation through a secure multi-party computing protocol to generate an initial global model, and prediction distribution of the global model is optimized in combination with knowledge distillation of differential privacy constraints; designing a gradient difference dynamic updating trigger mechanism of noise robustness, smoothing noise interference through a sliding window mean value, and realizing adaptive threshold calibration through linkage model performance verification; and light-weight deployment real-time reasoning is realized based on redundant edge pruning of confidence and 8-bit symmetric quantization. On the premise of protecting data privacy, the real-time performance and accuracy of health risk prediction are remarkably improved, and the method is suitable for a cross-institution medical care collaborative decision-making scene.

The invention relates to the technical field of data transfer and verification, in particular to a business data testing method and system and a storage medium. The method comprises the following steps: acquiring an original service data stream; deploying a supervision submission shadow processing sandbox based on the original business data flow to obtain a data isolation verification environment; configuring an alliance block chain server based on the original service data flow, and constructing a distributed data processing traceability platform; based on the distributed data processing traceability platform, performing data processing proof generation by using the original service data stream to obtain a data processing hash chain; configuring differential privacy protection processor parameters based on the data isolation verification environment to obtain a privacy protection test data generator; a privacy protection test data generator is trained to generate a synthetic test data set. By combining the alliance block chain and the differential privacy technology, a high-credibility, automatic and traceable data testing system is constructed, and the data security, the verification efficiency and the system consistency are remarkably improved.

A federated mining method for multimodal data based on multiple security policies includes: a federated learning framework is used as a data mining model for distributed data mining; a multiple authentication mechanism is designed; a local edge node is selected to participate in a federated computation to obtain authenticated local edge node and aggregate a dataset; a multimodal data fusion and a multimodal data classification are performed on the dataset by a generalized multimodal data feature fusion model based on a multi-headed attention mechanism; and an adaptive perturbation mechanism based on cyclic correlation analysis and differential privacy is designed to add noise round by round and dynamically. A federated mining method for multimodal data is further provided.

The invention discloses a data security and privacy management method and system based on a block chain technology, and the method comprises the following steps: S1, collecting original data in a target system, and carrying out the preprocessing of the original data; s2, constructing an access control strategy set; s3, converting into a structured strategy configuration file, and deploying a smart contract; s4, matching and judging the access request and a preset strategy by the intelligent contract; s5, judging whether the access request is legal or not; s6, performing structured processing on the access log, and writing the access log into a block chain account book; s7, intelligently selecting an encryption mechanism, and providing access return data to the access party; and S8, carrying out modeling prediction on the access mode by adopting a federal gradient lifting tree algorithm. The block chain intelligent contract and the federal gradient boosting tree algorithm are fused, differential privacy, zero-knowledge proof and homomorphic encryption technologies are combined, data intelligence and refined privacy protection are realized, and the method has the advantages of high credibility, high intelligence and high traceability.

The invention relates to a security protection method, system and device for data sharing and exchange and a medium. The method comprises the steps that original data are loaded, sensitive information in the loaded data is recognized, and the original data are classified according to the category of the sensitive information; obtaining user demand information and a corresponding target scene condition, adding noise to the sensitive information through a differential privacy algorithm according to the user demand information and the target scene condition, and generating privacy protection data; transmitting and encrypting the privacy protection data to obtain corresponding encrypted data, transmitting the encrypted data through a corresponding encrypted channel, recording an operation log in a data transmission process, and embedding an invisible watermark to track the use condition of the encrypted data; and analyzing the privacy protection requirement of the target scene condition, dynamically adjusting the privacy protection parameter of the encrypted data according to a preset use specification, and recovering the encrypted data when the life cycle of the encrypted data is ended. The method and the device have the effect of improving the data sharing security.

The present invention provides a privacy-enhanced structured data simulation and generation method and system. The method comprises: step 1, a data conversion phase: performing normalization preprocessing on data; step 2, a probabilistic graphical model construction phase: on the basis of a Bayesian form, constructing the posterior distribution for variational inference from the data having undergone normalization preprocessing in step 1, obtaining an association relationship between the features of structured data by using a Stein variational gradient descent method, and when introducing differential privacy noise, using a Monte Carlo estimation algorithm to automatically obtain the amount of noise that needs to be added for each update step; and step 3: a data generation phase: using the association relationship obtained in step 2 as a metric set to generate more accurate simulation data than real data. The beneficial effect of the present invention is that: the method of the present invention avoids gradient clipping when applying DP-SGD, thereby avoiding the selection of clipping parameters and reducing the adverse effects of gradient clipping on an inference process.

The invention relates to the technical field of information, and discloses a ciphertext execution exception handling method and system for a privacy computing algorithm, which quantifies the data query sensitivity of each tenant by collecting differential privacy query parameters of a plurality of tenants, and performs time sequence analysis on the sensitivity quantification result by using an ARIMA model to predict the future trend of the tenant, so as to improve the accuracy of the ciphertext execution exception handling of the privacy computing algorithm. The residual error between the actual sensitivity and the predicted sensitivity is calculated to identify abnormal tenants; a sliding window method is adopted to extract statistical features of the abnormal tenants to construct a feature set, and clustering analysis is carried out on the feature set to identify abnormal modes and abnormal fingerprint feature clusters corresponding to the abnormal tenants; and training a model based on the positive and abnormal fingerprint feature clusters of the abnormal tenants, and quantifying the influence degree of each abnormal mode on the privacy protection level of the abnormal tenants by using the model, so as to adjust and execute the privacy budget allocation proportion of each abnormal tenant according to the obtained quantitative analysis result. And therefore, the exception in the ciphertext execution of the privacy calculation algorithm can be effectively processed.

The invention discloses a medical whole-course intelligent management system based on a large model, and belongs to the technical field of large models. Comprising a multi-modal data acquisition module, a privacy calculation preprocessing module, a dynamic knowledge enhancement module, a time sequence data analysis module, an intelligent decision engine module, a multidisciplinary collaboration module, a patient interaction platform module, a dynamic intervention feedback module and a system security center module. The cross-mechanism data security sharing is realized, and the compliance of sensitive information processing is also ensured; a two-channel medical knowledge base is constructed, authoritative guidelines can be synchronized, newest clinical research data can be analyzed in real time, the knowledge base is kept in the newest state all the time, and the frontier scientific basis is provided for clinical decisions; dynamic modeling and trend prediction are carried out on long-term monitoring data of a patient by adopting a hybrid neural network model, and potential health risks and development trends can be identified more accurately.

The invention relates to the technical field of wireless communication networks, in particular to a vehicle ad hoc network based on intelligent networked vehicle WiFi and a V2V interaction method, a vehicle motion state and network load data are acquired through a vehicle-mounted sensor, AP / STA modes are dynamically switched by combining a deep Q network (DQN) model and an epsilon-greedy strategy, and an adaptive topological structure is constructed. And expanding OLSR protocol message embedding path delay fields, calculating and optimizing multi-hop routing selection in combination with pheromone gradient, and realizing message priority classification and differential privacy protection through a federated learning model. And dynamically adjusting an AP / STA proportion threshold based on a particle swarm algorithm, and triggering node role reconfiguration to balance network load. According to the method, the existing WiFi hardware is reused, the deployment cost is reduced, cross-layer routing optimization and a distributed decision mechanism are combined, and the network real-time performance, the expansibility and the group cooperation capability in a dynamic scene are improved.

The invention relates to the technical field of data processing, and discloses a risk prediction method and system based on dynamic aggregation and privacy protection. The method comprises the steps that multi-source heterogeneous data are collected and subjected to standardization processing; constructing a local risk association graph containing nodes, edges and topological attributes; inputting the map into a federated map neural network, and performing calculation through a local difference privacy mechanism to obtain fusion risk representation; and performing multi-level feature extraction and grade division on the risk representation to generate a comprehensive risk score and an early warning signal. Dynamic aggregation of multi-source heterogeneous data is achieved, data collaboration among multiple mechanisms is achieved while data privacy is protected, the accuracy and interpretability of risk prediction are improved through multi-level feature extraction, cross-mechanism risk knowledge fusion can be conducted under the condition that original data are not shared, and the risk prediction efficiency is improved. And the real-time performance and the accuracy of a risk prediction result can be ensured.

The invention relates to the technical field of pension information management, and discloses a cloud-based pension information management system, which comprises a health data acquisition module, a privacy protection processing module, a risk prediction module, a resource scheduling module and a comprehensive service module. The health data acquisition module synchronously acquires data in a multi-mode manner and encrypts and uploads the data by means of wearable equipment and an Internet of Things sensor; the privacy protection processing module guarantees data security by using federated learning and differential privacy encryption; the risk prediction module accurately predicts health risks based on a graph neural network; the resource scheduling module optimizes resource configuration through deep reinforcement learning; and the comprehensive service module generates a personalized service chain according to the knowledge graph. The system solves many problems of traditional pension information management, realizes multi-dimensional data real-time acquisition and privacy protection, accurately predicts health risks, efficiently schedules resources, provides personalized services, improves the pension service quality and the life quality of old people, and promotes the intelligent development of the pension industry.

The invention discloses a hierarchical collaborative management method and system for a virtual power plant based on multi-modal deep learning, and the method comprises the steps: constructing a four-dimensional data collection system, and achieving privacy enhancement preprocessing through federated learning and a differential privacy technology; a Bi-LSTM and a heterogeneous graph neural network are adopted to construct a three-mode deep fusion model, the weight is dynamically adjusted in combination with an environment-user dual-drive attention mechanism, and the load prediction precision and the space resource utilization rate are improved; a multi-target scheduling strategy is generated based on a five-dimensional target function and an improved DDPG algorithm, and physical feasibility is ensured through digital twinborn pre-verification; efficient execution and excitation transparency are realized through edge layer FPGA + NPU hardware acceleration and block chain evidence storage; and constructing a user participation ecology by using a natural language interaction strategy engine and a stepped incentive mechanism. The power grid economy, the equipment reliability and the user participation degree are remarkably improved, and intelligent upgrading of the virtual power plant is promoted.

The invention discloses a user data intelligent protection method and system based on differential privacy, and belongs to the technical field of data protection, and the method specifically comprises the steps: collecting and preprocessing user health data; setting a dynamic privacy budget according to the data sensitivity, the analysis purpose and the privacy demand, and adding random noise to generate disturbance data by adopting a multi-strategy fused data disturbance strategy; statistical analysis is carried out, and an improved differential privacy algorithm is used to assess the privacy leakage risk, the protection degree and the data availability; based on an evaluation result, a privacy budget and a disturbance strategy are iteratively adjusted through a feedback loop mechanism; the health data analysis method provided by the invention not only can protect user privacy, but also can ensure data availability.

The invention relates to the technical field of agricultural data security, and discloses an agricultural data security sharing method based on federated learning, through cooperative operation of a user terminal module, a data classification module, a federated learning module, an agricultural cloud database and a security protection module, a user terminal encrypts, uploads, decrypts and feeds back agricultural data through an edge computing device; the data classification module performs multi-level label labeling on data of planting industry, animal husbandry and the like by using a differential privacy technology; the federated learning module is combined with homomorphic encryption to realize local model training and parameter aggregation; the agricultural cloud database stores encrypted knowledge assets by using an IPFS technology; the security protection module integrates block chain evidence storage and a dynamic access control mechanism, the workflow comprises data encryption uploading, privacy protection classification, federated model training, solution generation and security feedback, and by adopting a federated normal form of a data immobile model, the security protection module can effectively protect the security of the original data on the premise of ensuring that the original data is not out of the domain. And value mining and safe sharing of cross-regional agricultural data are realized.

The invention provides an investment decision-making system and method based on federated learning and multi-modal risk control, and the system comprises a data collection layer which is used for integrating satellite remote sensing data, supply chain Internet of Things data and financial market data; the federal processing layer is used for constructing a distributed training framework based on a FedAvg algorithm and realizing cross-mechanism data fusion and global model parameter updating through homomorphic encryption and differential privacy; the intelligent decision-making layer adopts a TD3 reinforcement learning engine and adopts a dynamic gating mechanism to adjust fusion weights of texts, time sequences and image modalities in real time; and the execution monitoring layer is used for detecting transaction execution deviation in real time through a digital twinning technology, realizing abnormal response by adopting a CUSUM algorithm, and dynamically adjusting a federal learning rate through Bayesian optimization. The key technical defects of data islands, tail risk response delay, insufficient cross-market linkage analysis and the like of a traditional investment system are overcome through an innovatively designed federated learning architecture and a time-varying risk control model.

The invention discloses a built environment and traffic accident causal effect analysis method based on bi-robust learning, and belongs to the field of intelligent traffic safety and traffic accident risk assessment. By integrating satellite remote sensing, interest point distribution and traffic accident data, a spatial-temporal feature enhanced environment index system is constructed; realizing cross-regional data collaborative analysis by adopting federal learning, and guaranteeing data compliance in combination with differential privacy encryption; a generative adversarial network is introduced to generate a high-fidelity anti-fact scene, and a dual robust learning framework is coupled to accurately quantify a condition average processing effect of a built environment on accident risks. According to the scheme, a technical tool with both theoretical preciseness and practical feasibility is provided for urban traffic planning and safety management, and a scientific basis is provided for optimizing and building environment design and reducing traffic accident risks through a multi-source data fusion and privacy cooperation mechanism; and the blank in the aspects of causal robustness, insufficient anti-fact generation precision, privacy protection and multi-source data fusion in the prior art is filled.

The invention relates to the technical field of AI interaction, and discloses an AI multi-modal emotion interaction memory terminal, which realizes microsecond-level synchronization of voice, facial expression and text data through a multi-thread acquisition engine, dynamically allocates each modal weight by adopting a multi-head cross attention mechanism, and adaptively adjusts modal importance based on a conversation context hidden state; when the cross-modal confidence difference exceeds a threshold value, a gating LSTM conflict resolution module is activated, and the multi-source data collaboration problem is solved; the emotional memory modeling constructs an emotional state transition topology based on a graph convolutional network, protects user privacy in combination with a differential privacy mechanism, and realizes associated event storage of millisecond backtracking of short-term memory and long-term memory. The technology integrates multi-modal dynamic perception, privacy security calculation and adaptive learning ability, significantly improves the real-time performance and personification degree of emotion interaction, and can be applied to the fields of intelligent customer service, emotion accompanying, health monitoring and the like.

An AI-driven system for personalized advertising in real time, where: ◯ an analytics unit to monitor and analyze user behavior in real time across multiple digital platforms such as websites, mobile applications, social media, and smart devices; the unit collects data on user engagement, browsing patterns, time spent, and content preferences to enable targeted, personalized advertising; ◯ a prediction module that predicts preferences and interests of a user, operatively connected to the user interaction data collection unit, wherein the prediction module uses machine learning models such as deep learning, recurrent neural networks (RNNs) and transformer-based architectures to predict interests of users based on historical interactions and inferred preferences; ◯ an emotion and sentiment analysis unit that assesses the user's mood in real time through computer vision, natural language processing (NLP) and voice analysis, whereby the analysis of facial expressions, voice pitch and linguistic mood is used to determine emotional states and receptivity to advertising content; ◯ an embodiment of a context awareness component in operational communication with the emotion analysis and mood unit, in some cases further augmented by various environmental and situational data such as the device type and its physical location, date and time, and the content processed in the device, processing methods, etc., in establishing adaptability and automatic ad placement to be as relevant as possible to the user and their status as prescribed; ◯ Use reinforcement learning algorithms and generative AI models to drive advertising with personalization engines. Creative elements, messaging, and presentations are dynamically adjusted in real time based on user responses to ensure advertising is personalized and always optimized for best performance; o a privacy-focused AI system with federated learning, differential privacy methods, and on-device AI processing to reduce targeted advertising while complying with international data protection laws such as the General Data Protection Regulation (GDPR) and California consumer privacy laws; o an operationally adapted ad delivery mechanism to engage with real-time bidding (RTB) networks, programmatic advertising exchanges and demand-side platforms (DSPs) and place advertisements through digital advertising networks, which guarantees the delivery of tailored advertising to the most relevant audience in real time; and o a contextual feedback loop in which the machine learning models used in the preference and interest prediction module and in advertising personalization The engine is continuously updated to reflect the latest user engagement data, improving personalization over time and optimizing advertising performance.

The invention discloses a dynamic power distribution network deduction method based on graph neural network perception and a privacy protection type Transform model. The method comprises the steps of collecting source load storage historical data in a power distribution network and performing standardization processing; performing time sequence segmentation; establishing a dynamic topological structure model of the power distribution network; extracting spatial features of the dynamic topological structure of the power distribution network by using a graph neural network; the spatial features are converted into time sequence data, the time sequence data are combined with time sequence data obtained after time sequence segmentation, and a source network load storage time sequence data set is established; differential privacy processing is carried out; training a Transform model, in the training process, using a differential privacy optimization technology, adding noise to the gradient when the gradient is updated each time, and finally obtaining a trained DP-Transform model; and inputting a specific scene, and deducing the state of the power distribution network by using a DP-Transform model. According to the method, the interpretability of deducing the state of the power distribution network by using the Transform is improved, so that the Transform model can fully receive the graph structure input of the network topology structure of the power distribution network, and privacy protection on user information in the deducing process is considered.

The invention relates to the technical field of data processing and privacy protection, in particular to a data desensitization and integrity verification method and system based on a differential privacy algorithm, and aims to realize collaborative optimization of data privacy protection and integrity guarantee. Determining a total privacy budget by calculating global sensitivity, a data leakage risk coefficient and independent and joint data privacy information amount, dynamically allocating the privacy budget for a single-task or multi-task scene, and injecting noise based on an allocation result to realize differential privacy desensitization; meanwhile, a verification voucher is generated by using a hash function, a digital signature and an alliance chain, and the data integrity is ensured through signature verification, data consistency verification and block chain verification. The system comprises a data application processing module, a global sensitivity calculation module, a desensitization core processing module, a verification voucher generation module and an integrity verification module, is suitable for government affairs, enterprises and other scenes with high requirements for data privacy and integrity, and effectively improves the safety and reliability of data processing.

The invention relates to a federated learning system based on multi-key homomorphic encryption and adaptive differential privacy, and belongs to the technical field of privacy computing. According to the system, on the premise that no trusted third party exists, a multi-client collaborative key generation and threshold decryption mechanism is achieved, it is ensured that model parameters are always in an encrypted state in the aggregation process, and leakage of a single node is prevented. By introducing a parameter sensitivity analysis and selective encryption strategy, the system only encrypts high-risk parameters, and the encryption burden is effectively reduced. Meanwhile, in combination with an adaptive privacy budget allocation mechanism, the system dynamically adjusts noise intensity according to a model training state, and model performance and convergence speed are maintained while privacy protection capability is improved. According to the method, high robustness and collusion resistance are realized, stable operation under the condition that part of clients are offline is supported, and the method is suitable for application scenes such as medical treatment and finance with high data sensitivity and strict performance requirements.

The invention discloses a privacy protection system based on federated learning, and relates to the technical field of privacy protection. The system comprises a distributed participating node cluster, wherein each participating node is configured with a local model training unit and a privacy protection module; the coordination server is connected with each participating node through a secure communication layer and comprises a model aggregation module and a dynamic trust evaluation module; the global model distribution channel is used for broadcasting the encrypted global model parameters to participating nodes; and the privacy protection module is integrated in a local participating node and comprises a homomorphic encryption engine and a local differential privacy injector. According to the method, privacy protection strength improvement, model utility optimization, system efficiency breakthrough and security boundary expansion are synchronously achieved under a federated learning framework, and an industrial-grade solution is provided for cross-domain data collaborative learning.

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.