35 results about "Model aggregation" patented technology

Embodiments of the present application disclose a data clustering and summarizing method and device, equipment and a storage medium. The method comprises: obtaining a target suggestion dataset corresponding to a suggestion type problem insight task; performing vectorization clustering on the target suggestion dataset to obtain a target suggestion clustering result corresponding to the target suggestion dataset; using a target large model to summarize the same type of target suggestion data in the target suggestion dataset based on the target suggestion clustering result, to obtain target summary description data corresponding to each suggestion category in the target suggestion dataset. In the clustering insight scene of the suggestion type problem, through the vectorization clustering and large model summarizing manner, not only the artificial cost required for the suggestion data clustering and summarizing can be saved, and the efficiency of the suggestion type problem sorting can be improved, but also the same type of suggestion data can be expressed and summarized according to the suggestion clustering result, so that the user can efficiently and intuitively understand the key points of each suggestion category in the suggestion clustering result.

This invention relates to a self-learning method and system for home environment regulation based on multimodal feedback and scene awareness. The method collects the user's physiological characteristics and body movement signals through non-contact sensing devices to identify the user's current activity scene. The system integrates the user's explicit intervention operations with implicit comfort feedback based on physiological stability, dynamically assigning fusion weights for explicit and implicit feedback according to the activity scene to generate target feedback labels. An online incremental learning algorithm is used to update the local environmental control model, and a privacy-preserving model aggregation is performed in the cloud through a federated learning mechanism. This invention solves the problems of traditional home control relying on single commands and lacking physiological feedback mechanisms, achieving intelligent closed-loop control that balances user privacy and personalized comfort.

Provided are a communication method, a terminal device, and a network device. The communication method is applied to a machine learning model and comprises: a terminal device receiving a global model parameter related to a first model sent by a network device; the terminal device sending a first parameter to the network device, wherein the first parameter is determined on the basis of the global model parameter; the terminal device receiving a plurality of candidate parameters sent by the network device; and on the basis of the plurality of candidate parameters, the terminal device determining a local model aggregation parameter that is to be sent to the network device, wherein the terminal device is one of a plurality of terminal devices, the plurality of candidate parameters are determined on the basis of a plurality of first parameters sent by the plurality of terminal devices, and a plurality of local model aggregation parameters of the plurality of terminal devices are used for training the first model.

The application relates to a safe and efficient model co-construction method, and belongs to the field of artificial intelligence. The method comprises the following steps: regional division: each edge server divides a responsible management region according to the range coverage capacity thereof; system initialization: a global model and a key generation are initialized; local model training: model updating is calculated according to local data of equipment, and gradient information is disturbed and returned; edge security robust aggregation: a communication-efficient security-enhanced aggregation protocol is designed to support the implementation of an asynchronous grouping robust aggregation algorithm based on disturbed gradients; cloud global model aggregation: local model aggregation results returned by each edge server are received, and a federal average algorithm is executed to perform global model aggregation. The application can effectively improve the robustness and security of a global model under the condition that a client exists device heterogeneity and resource limitation.

A method and system are provided for Random Sampling Consensus Joint (RSCFed) learning in a non-IID setting. The method includes: randomly sampling local clients; initializing the current global model by assigning it to the randomly sampled local clients at the start of a synchronization round; training the randomly sampled local clients locally; collecting local models from the randomly sampled local clients and performing distance-reweighted model aggregation (DMA) on the collected local models to obtain sub-consensus models; repeating the above steps multiple times to obtain a set of sub-consensus models; and aggregating a new model based on the sub-consensus models as the next global model.

This invention discloses a federated learning method that enhances privacy and robustness, relating to the fields of privacy protection and federated learning technology. The technical solution includes the following steps: S10, system initialization; S20, encrypting the local model; S30, calculating the weights of the local gradients; S40, model aggregation; S50, local model update; S60, reward allocation. This method employs fully homomorphic encryption to encrypt the participants' local gradients, protecting their privacy; and introduces an attention mechanism to detect and suppress malicious gradients in the encrypted state, effectively resisting poisoning attacks and improving the robustness of the global model. Simultaneously, it uses a consortium blockchain to record the federated learning training process, calculates contribution values ​​based on the participants' attention weights, and allocates rewards or deducts deposits accordingly to incentivize honest participants to provide high-quality data over the long term.

The application discloses a kind of federated dynamic weighted contribution evaluation methods and systems for data privacy protection, method includes: each participant is locally trained to obtain model parameter update using private data, and the performance gain value is encrypted using homomorphic encryption algorithm to generate encrypted gain ciphertext;Central server receives the privacy protection update and encrypted gain ciphertext of all participants, generates contribution degree vector;Server is based on contribution degree vector and combines historical contribution record dynamically calculates the weighting coefficient of next round aggregation, generates weighting coefficient vector, obtains weighted update ciphertext;Server carries out homomorphic aggregation to obtain global model update ciphertext using weighted update ciphertext, and each participant uses private key to obtain global model update and updates local model.Using the embodiment of the application, the full-process privacy protection of participant contribution evaluation and model aggregation can be realized, and the convergence efficiency and performance stability of global model are effectively improved.

This invention provides a federated learning optimization method and system for imbalanced datasets. In each round of communication, the server sends a global model, data distribution divergence, and a global set of difficult categories to the client. The client adjusts its learning rate accordingly and adds weights for difficult categories to the loss function. After completing local training, the client uploads its local model and local state vector. The server aggregates the state vectors of each client, calculates the new set of global difficult categories and the data distribution divergence of each client, and adjusts the model aggregation weights based on factors such as whether the client's divergence is abnormal and the number of samples in the difficult categories. When aggregating classification layer parameters, the average loss of each client in each category is further combined to achieve refined model optimization for class imbalance and data distribution differences.

Provided in the present invention are a decentralized hierarchical federated learning method and system, and an edge server. The method comprises: a plurality of edge servers respectively receiving gradient ciphertexts sent by different participant clients; the plurality of edge servers aggregating the received gradient ciphertexts on the basis of a secure multi-party computation protocol, so as to obtain an aggregated model ciphertext, wherein the secure multi-party computation protocol refers, in a mutually untrusted multi-user network, to a plurality of edge servers respectively holding different gradient ciphertexts jointly performing computation to obtain an aggregated model ciphertext based on these gradient ciphertexts, and each edge server only partially obtaining data in the aggregated model ciphertext, and not leaking its own gradient ciphertext to other edge servers; and the edge servers issuing the aggregated model ciphertext to the participant clients. By using the solution of the present invention, the security of model aggregation can be improved.

This invention discloses an attention-integrated multimodal federated learning method and system for risk monitoring. On the client side, an attention-integrated hybrid early fusion module dynamically injects multimodal features into globally shared latent variables through attention mechanisms and entropy-based gating mechanisms. On the server side, a plug-and-play dual-track federated aggregation strategy decouples state aggregation from model aggregation. The state track employs robust filtering based on the geometric median, while the model track supports algorithm-independent parameter updates. Between communication rounds, a neighborhood-weighted personalized reference mechanism constructs a gradient-based semantic topology to balance global consistency and local task preferences. This invention simultaneously addresses the technical challenges of modality incompleteness, data statistical heterogeneity, and Byzantine adversarial robustness in multimodal federated learning, thus providing an efficient and robust collaborative learning solution for disaster risk perception under privacy-preserving conditions.

The application discloses a kind of sharding blockchains federated learning methods for AI big model, belong to information security field, the application designs the scalable big model training architecture based on sharding blockchains and federated learning, with the aid of sharding blockchains technology, the scalability of federated learning system is guaranteed from architecture level.Based on the multi-piece collaborative model aggregation scheme SP of four-pipeline two-stage commitment, which includes multi-signature aggregation protocol MASign, pipeline mode intra-chip consensus protocol MulPipe-BFT based on improved BLS multi-signature, and inter-chip transaction processing protocol QuadPipe-Sharding based on four-pipeline mode multi-signature aggregation.The application designs a federated learning endogenous security enhancement technology composed of uniform noise adding technology, phased encryption technology and Bipole double filter, which resists data recovery attacks, free riding attacks and poisoning attacks and other risks faced during federated learning training process.

The application provides a federated learning method and system based on data distribution similarity fuzzy clustering, belongs to the technical field of federated learning, and broadcasts all cluster models; loss values of the cluster models are calculated, N clusters are selected as associated clusters of a client according to the loss values, and importance of the clusters is evaluated, wherein the value of N is a parameter determined in advance according to the number of clusters and the number of clients; a client model is initialized, and local training is performed; cluster model aggregation weights are updated according to a client sample size and an association degree between the client and the clusters, and an updated cluster model is obtained. By associating one user to multiple clusters, the application effectively improves the problem of mixed distribution of multiple data, can make the network model better converge and generalize, and has more excellent individualization ability.

The application discloses a kind of based on multi-attribute ordering's federal learning participant trust evaluation method, equipment and storage medium, belong to cyberspace security field;Its method includes: aggregation server issues federal learning task, and defines initial global model distribution to each participant;Participant trains local model based on local data, and sends to aggregation server;Aggregation server collects the behavior information interacted with participant, and the recommended trust information of recommendation server, and the behavior information and recommended trust information are used as trust evidence, according to trust evidence, the trust attribute comprehensive value of participant is calculated, and after updating trust information based on trust attribute comprehensive value, the local model uploaded by the participant ranking in front is selected to participate in model aggregation, after aggregation, the aggregated model is sent to all participants;Equipment and storage medium are used to realize the method;The application improves federal learning performance and model precision, and has important application value in federal learning field.

The application discloses a kind of based on federated learning's intelligent adjustment model distributed training method of network-connected vehicle, it is related to artificial intelligence and distributed computing field.The application includes: through high-definition flexible camera real-time capture dynamic operation component image, and realize the standardization pretreatment of adjustment image data;Based on each participant node definition sample distribution rule, construct the distributed training data set with same label space and non-independent and identically distributed characteristics;Initialize YOLOv5 global weight parameter, construct the distributed model training framework based on federated learning architecture.Through operation site multidimensional image acquisition, unbalanced data set construction, distributed model framework construction, participant local model training, server model aggregation update and model multidimensional index test, realize the cross-subject collaborative use of adjustment operation data under the premise of privacy security, greatly improve the generalization ability and fault recognition precision of intelligent adjustment model, meet the intelligent operation and maintenance demand of network-connected vehicle under complex working condition.

The application provides a federal incremental learning method in an automatic variable load system of an air separation device, which is reasonable in structural design. The incremental learning is introduced in the federal learning, which helps each client to train data when there are only a small number of classes initially, and then gradually adds learning when the classes are increased. When facing new tasks and data, retraining is no longer needed. The first module of the application is a federal incremental learning module, which extracts the same number of samples from each client for pre-training. Considering the cost loss of retraining, the iCaRL strategy is added to the traditional federal learning framework, which can cope with dynamic changes of training tasks and mutual privacy of user data. The second module of the application is a double attention mechanism. Considering the influence of large sample clients on the final model training result and the difficulty of catching up with the feature imbalance during model aggregation, a channel attention neural network model is designed for the client, and a federal aggregation algorithm based on the feature attention mechanism is designed for the global model.

This invention provides an energy prediction system based on quantum particle swarm optimization (QPSO) federated spatiotemporal coupling, comprising an edge sensing layer, a federated computing layer, a quantum optimization layer, and a spatiotemporal prediction layer. The edge sensing layer collects operational data from distributed energy nodes and performs data cleaning and feature encoding. The federated computing layer jointly models the spatial relationships between energy nodes and the temporal evolution characteristics of operational data, generating model update information. The quantum optimization layer triggers a federated aggregation process when preset conditions are met and updates the model aggregation weights based on the QPSO mechanism, generating global model parameters. The spatiotemporal prediction layer predicts and analyzes the operating status of the energy system based on the global model parameters and outputs risk assessment information. This invention achieves multi-node collaborative modeling without centralizing raw energy data, which is beneficial for improving the adaptability and application feasibility of the energy prediction system in complex operating scenarios.

The application discloses a provable incentive mechanism and a reward distribution method for federated learning, and through the introduction of a trusted execution environment and remote attestation technology, the provable execution of the federated learning incentive mechanism is realized under the threat model that the server may not be honest, the whole process of contribution evaluation, reward distribution and model aggregation is ensured to be transparent and verifiable to the client, and the dishonest behavior of the server in tampering with the reward distribution is effectively restricted; meanwhile, through the design of a contribution-aware reward distribution algorithm based on a reverse auction game, the client reward is positively correlated with the real contribution degree, the high-quality client is effectively encouraged to continuously participate while the server budget constraint is considered; in addition, the core function is deployed in the trusted execution environment by using a modular architecture, the internal and external interaction and the code size are minimized, the security attack surface is reduced, and the stability, maintainability and scalability of the system are improved, so that a fair, trustworthy and efficient federated learning incentive ecosystem is constructed.

Embodiments of the present application provide a federated learning data processing method and device, a storage medium and a computer program product, and belong to the technical field of machine learning. The method comprises: distributing an initial global model to a plurality of clients; receiving model gradient information uploaded by each client after the initial global model is trained based on a local data set of the client; randomly generating a plurality of candidate clustering schemes, wherein each candidate clustering scheme corresponds to a grouping manner of dividing all clients into different clusters; determining a client set in each cluster according to the candidate clustering scheme, calculating an aggregated gradient of each cluster, and determining an evaluation index of the candidate clustering scheme based on the aggregated gradient of all clusters and an independent data set for model evaluation; determining a target user clustering scheme from the plurality of candidate clustering schemes according to the evaluation index; grouping the clients based on the target user clustering scheme, and performing model aggregation training for each group respectively to obtain a final model corresponding to each group.

The application discloses a harmonic reducer fault diagnosis method and system based on federal prototype domain generalization under unknown working conditions, belongs to the technical field of machine fault diagnosis, and solves the problem of federal domain generalization under the privacy distribution of multi-user industrial robot harmonic reducer data islands. The method comprises the following steps: firstly, an adaptive normalized hyperspherical surface module is constructed to enhance the ability of a user local model to extract local data domain invariant features. Secondly, a local model training mechanism guided by a prior fixed prototype is designed to improve the consistency of domain invariant features among all users, so that the generalization performance of the global model to unseen working condition data is enhanced. Finally, in order to reduce the negative influence caused by the data distribution difference caused by the imbalance of data categories among users, a personalized model aggregation strategy based on dynamic model decoupling is proposed. The application is suitable for harmonic reducer fault diagnosis under unknown working conditions.

The application relates to the technical field of federal learning and industrial data security, and discloses a federal learning model cooperative training and privacy protection system for industrial big data, which comprises an industrial data access layer, a local federal training layer, a federal cooperative scheduling layer, a privacy protection reinforcement layer, a model aggregation verification layer and a model deployment application layer, each layer is cooperatively linked from top to bottom, the privacy security and efficient cooperative training of the industrial big data are realized, the application has strong adaptability, can effectively adapt to the characteristics of the industrial big data, such as time sequence, heterogeneity and massiveness, a time sequence weight factor is introduced through an improved federal learning algorithm, personalized training hyperparameter configuration is combined, the model training precision and training efficiency are improved, and the problem of poor adaptability of the existing system is solved, the application has high privacy protection strength, high cooperative efficiency, reliable aggregation results, and strong practicality and expandability.

A model training unit (120) trains a target model using learning data and generates a trained model. An inverse training unit (130) inversely trains the target model using the learning data and generates an inversely trained model. A model aggregation unit (140) aggregates the trained model and the inversely trained model to generate a pre-trained model.

The application discloses a federal learning model compression method, a user terminal, a server and a system, and belongs to the technical field of communication. The method comprises the following steps: acquiring a local model corresponding to at least one target application; determining a compression strategy of the local model based on a performance index of the target application; performing compression processing on the local model based on the compression strategy to obtain a local compressed model and a compression identifier corresponding to the local compressed model; uploading the local compressed model and the compression identifier to a server for model aggregation processing, acquiring global model parameters issued by the server; and updating the local model based on the global model parameters. In this way, according to the performance index of the application, a suitable compression strategy is selected for model compression, the communication cost can be reduced on the basis that the federal learning model is deployed to multiple devices, the accuracy of the federal learning model parameters is maximized, and the experience of users is significantly improved.

This invention discloses a backdoor defense method for federated learning based on gradient filtering and weight adjustment, relating to the field of federated learning security. The method first identifies and eliminates potential anomalous updates through gradient adaptive sparsity, cluster analysis, and anomaly detection to reduce redundant information and lower the risk of malicious attacks. Then, during the aggregation process, the weights of each client are adaptively adjusted based on factors such as gradient similarity, optimizing the model aggregation process and ensuring that the contributions of benign clients are effectively utilized while mitigating the impact of malicious clients. Compared with existing defense algorithms, this invention can effectively resist various backdoor attacks without significantly sacrificing model accuracy. It exhibits superior defense performance, robustness, and low computational complexity in different data environments and can be widely applied to various federated learning systems.

This application discloses a method and system for evaluating the contributions of multi-dimensional participants, as well as a model aggregation and optimization method. The evaluation method includes: distributing an initialized global model to the multi-dimensional participants; each participant performing iterative training based on local data; determining the comprehensive accuracy improvement factor for each client based on the local model information after each iteration; then adjusting the weights of the magnitude, direction, and uniqueness dimensions based on the obtained model accuracy change rate; obtaining a dynamically adjusted comprehensive accuracy improvement factor based on the adjusted weights; dynamically aggregating the locally trained model parameters to obtain updated global model parameters, which are then distributed to the multi-dimensional participants for local training; stopping training when the stopping condition is met and obtaining the factor for the corresponding round; and simultaneously determining the model contribution information for each participant by combining historical decay cumulative contribution. This method can efficiently and accurately obtain the contribution information of each participant.

The application discloses a kind of distributed integrated learning methods based on blockchain, based on the three-layer blockchain structure designed by mini-block, integrated block and key block, the miner in network is trained on the training set after with replacement sampling by consensus protocol, aggregate model from other miners, finally the information of base model and integrated model is recorded in blockchain, so that the whole process of model training, model integration, model evaluation is integrated into blockchain consensus, so that the whole integrated learning process is automatically executed in blockchain network.Therefore, the application can improve the utilization rate of computing power in the blockchain network based on the useful work of machine learning proof mechanism, while avoiding the introduction of central node in the model aggregation process, maximizing the degree of decentralization of blockchain.

The application discloses a layered federated reinforcement learning energy optimization method for a zero-carbon park, belongs to the technical field of energy optimization, and aims to solve problems of data privacy leakage in traditional centralized optimization and performance decline of a federated learning in processing non-independent and identically distributed data of heterogeneous enterprises, takes a data static model dynamic as a core, constructs a layered federated aggregation architecture of an enterprise layer, a grouping layer and a global layer, adopts a PPO reinforcement learning intelligent agent on the enterprise side, trains and encrypts model parameters based on local data and uploads the model parameters, divides enterprises into groups through JS divergence and K-means clustering on the park side, completes model aggregation through grouped federated averaging and global secondary fusion, and then updates and issues an optimization strategy, the application combines a multi-objective reward function of economic cost, equipment loss and carbon emission cost, realizes distributed autonomous optimization and global collaborative scheduling of the park, improves high-proportion new energy consumption capacity, and takes into account economic and zero-carbon targets.

This application provides a code analysis method, device, storage medium, and computer program product; the method includes: a first device acquiring multiple first code files corresponding to a first service, and performing code segmentation on the multiple first code files to obtain multiple second code files; the multiple first code files are used to implement the functional requirements corresponding to the first service; for each of the multiple second code files, the second code file and multiple case files are analyzed based on a first large model to obtain multiple analysis results, and the multiple analysis results are summarized based on the first large model to obtain a first report corresponding to the second code file; the multiple case files are used to test whether the functions implemented based on the multiple first code files meet the functional requirements corresponding to the first service; based on the first large model, the multiple first reports corresponding to the multiple second code files are summarized to obtain a second report; and the second report is output to a set first channel.

This application relates to a method, system, device, and medium for secure defense during self-supervised model training. The method includes: performing joint frequency and spatial preprocessing on a local training dataset with original labels to generate a preprocessed labeled training set; removing labels to generate an unlabeled dataset, and obtaining purified feature extraction layer parameters through self-supervised learning; fixing these parameters and minimizing the symmetric cross-entropy loss of the classification layer based on the labels to obtain basic model parameters; partitioning the dataset according to the forward propagation loss and obtaining the final model parameters and update gradients by combining them with a semi-supervised joint loss; generating federated aggregation control instructions based on the update gradients and sending them to the server to control global weighted aggregation, generating the next round of globally secure model parameters. This method can improve the security and robustness of self-supervised model training and ensure the reliability of model aggregation in federated learning.

The application 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, each participating node is configured with a local model training unit and a privacy protection module, a coordination server connected with each participating node through a secure communication layer, containing a model aggregation module and a dynamic trust evaluation module, a global model distribution channel for broadcasting encrypted global model parameters to participating nodes, and a privacy protection module integrated in the local participating node, containing a homomorphic encryption engine and a local differential privacy injector.The application synchronously achieves the improvement of privacy protection strength, the optimization of model utility, the breakthrough of system efficiency and the expansion of security boundary under the federated learning framework, and provides an industrial-level solution for cross-domain data collaborative learning.