474 results about "Model aggregation" patented technology

The invention discloses a multi-modal data real-time identification and cooperative processing system based on edge computing and federated learning. The multi-modal data real-time identification and cooperative processing system comprises a cloud center coordination node, a plurality of edge computing nodes, a cross-modal encryption engine, a federated learning controller and a model updating verification module. The cloud center coordination node executes federated learning model aggregation and dynamic task allocation, and generates a cross-modal encryption strategy; and the edge computing node is configured with a multi-modal data acquisition module, a local model training unit and a co-processing gateway to realize multi-modal data acquisition and local processing. The system encrypts vision, acoustics and text data by using differentiated algorithms such as spatial confusion, frequency domain permutation and homomorphic encryption; the federated learning controller carries out multi-modal feature fusion, hierarchical encryption and dynamic networking at the edge node; and the model updating verification module performs aggregation updating after ensuring parameter consistency by using secure multi-party calculation. According to the method, real-time processing and privacy protection of multi-modal data are realized, and the data co-processing efficiency is improved.

The invention relates to the technical field of agricultural information management, and particularly discloses an agricultural information management system and method based on a big data platform, and the method comprises the steps: firstly deploying a multi-source data collection module at an edge calculation node, and obtaining and standardizing the soil moisture content, meteorological environment and equipment operation data in real time; secondly, constructing a local dynamic irrigation strategy model, and realizing multi-objective optimization through a reinforcement learning algorithm; establishing a federated learning framework at the cloud, dynamically distributing node weights by adopting an attention mechanism, and realizing model aggregation of privacy protection in combination with secure multi-party computing; an optimal irrigation instruction is generated through a multi-source data fusion engine, and a three-level response exception handling mechanism is established; and finally, a closed-loop feedback system containing short-term incremental learning and long-term architecture optimization is formed. The corresponding management system comprises six functional modules, namely a data acquisition module, a local modeling module, a federated learning module, a real-time decision-making module, an abnormal monitoring module and a closed-loop optimization module.

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 meteorological monitoring method based on a GRU neural network and differential privacy federated learning. The meteorological monitoring method comprises the following steps: collecting local meteorological data to generate a meteorological time sequence data sequence; selecting a proper model structure according to the heterogeneity of a client, carrying out local meteorological monitoring model training based on a GRU neural network, and carrying out parameter updating on a local model through an LPDP-FedProx federal learning algorithm; the central server performs model aggregation and updating according to a global model weighted average aggregation algorithm after acquiring model parameters of the client; and the client receives the updated global model parameters and carries out the next round of local training until an optimal monitoring model is obtained. According to the method, the personalized differential privacy algorithm and the FedProx federated learning algorithm are combined, the LPDP-FedProx federated learning algorithm is provided, a federated learning framework capable of effectively processing heterogeneous data is constructed, the privacy of local data is effectively protected, and the security of federated learning is improved.

The invention relates to the technical field of distributed learning, in particular to a large model pre-training system based on distributed parallel processing. In the system, a data distribution layer collects a node resource state through a fragmentation module and generates a dynamic scheduling strategy; a computing resource layer configuration model initialization module and a strategy switching module support flexible switching of multiple modes such as tensor parallelism, data parallelism and assembly line parallelism; the network communication management layer is combined with topology perception and gradient compression technologies, so that the communication efficiency is improved; and the model aggregation layer realizes global parameter updating and training tuning under privacy protection through a security aggregation and optimization control mechanism. All layers of the system operate cooperatively, the calculation efficiency, the communication performance and the data security of large model pre-training can be effectively improved, and the method is suitable for model development and deployment in a large-scale heterogeneous calculation environment.

The invention discloses an AI oT-based cloud network side-end federated learning system, and relates to the technical field of cloud network side-end federated learning, comprising the combination of Internet of Things, cloud computing and artificial intelligence technology, and through intelligent cooperative computing of cloud-side-end three-layer computing resources, the cloud network side-end federated learning is realized. The AI oT-based cloud network side-end federated learning system further comprises a cloud module, a network module, an edge module and a terminal module, and has the advantages that local model aggregation is firstly carried out on edge nodes, a part of calculation tasks can be completed at an edge end, a model result subjected to primary processing and aggregation is uploaded to the cloud, and the cloud network side-end federated learning is realized. Different edge nodes and terminal devices have different computing capabilities, the cloud end dynamically distributes training tasks to match tasks with different complexity degrees with devices with corresponding computing capabilities, and the edge nodes with strong computing capabilities can undertake more complex model training sub-tasks. And the terminal equipment with relatively weak computing power is responsible for some simple tasks such as data preprocessing and the like.

The invention discloses a data privacy protection method based on multi-party security computing, which relates to the field of data security, and comprises the following steps: encrypting original data by adopting a Paillier semi-homomorphic encryption algorithm, computing a kernel matrix in an encryption state by utilizing homomorphic properties, training a support vector machine model in combination with a homomorphic sequence minimum optimization algorithm, and carrying out multi-party security computing. And differential privacy is introduced through a federated learning framework to carry out noise processing on model parameters, and finally model aggregation among distributed participants is realized by adopting a security summation protocol. According to the method, the data security and privacy can be ensured, a plurality of participants can cooperatively train a high-performance support vector machine model, and a privacy protection prediction service is provided for a user based on a global model obtained through training. In view of the problem of low calculation efficiency in multi-party security calculation in the prior art, the method effectively solves the data island dilemma and privacy disclosure risk faced by traditional machine learning, and is helpful to improve the efficiency of data sharing and analysis.

The invention discloses a multi-source threat intelligence privacy fusion processing method and system, and relates to the technical field of network security threat intelligence analysis and processing, and the method comprises the steps: collecting threat intelligence, and carrying out the cleaning, standardization and desensitization; establishing a threat index semantic model and aligning cross-modal features; the central coordination module completes model aggregation after organization node local training; and based on the aggregation model fusion intelligence, semantic association and attack link reconstruction are established, and a result is generated and fed back for optimization. The technical problems that in cross-organization fusion processing of multi-source heterogeneous threat intelligence, data formats are not uniform, and privacy protection and intelligence collaborative analysis contradictions exist, so that threat detection is not comprehensive and inaccurate, and accurate evaluation and effective management and control requirements are difficult to meet are solved. The technical effects that the multi-source heterogeneous threat intelligence is effectively fused on the premise of privacy protection, the comprehensiveness and accuracy of threat detection are improved, and the requirements for accurate assessment and effective management and control of threats in a cross-organization scene are met are achieved.

The invention belongs to the technical field of vehicle dynamics prediction, and discloses a vehicle data prediction method based on GRU and differential privacy federated learning, and the method comprises the following steps: a server side randomly initializes global GRU model parameters, and then carries out the training of a local GRU model; the client processes the trained local GRU model parameters through a differential privacy mechanism to generate a disturbance model; each client uploads the disturbance model parameters to a server side; carrying out global GRU model aggregation by adopting a weighted aggregation algorithm in federated learning; and the server side redistributes the aggregated global GRU model to each client side, and the global GRU model for accurately predicting the vehicle dynamics is obtained through multiple rounds of training and updating. According to the vehicle data prediction method based on GRU and differential privacy federated learning, through combination of a GRU network and a differential privacy mechanism, vehicle dynamics prediction in an Internet of Vehicles environment is realized, and privacy of a user is effectively protected.

Various systems and methods are presented regarding monitoring and controlling operation of a process. A visual representation of the process can be created based on a supermodel comprising models (representing one or more devices) and nodes (representing respective device variables and constraints). Further, the process can be represented by levels, wherein devices at each level can be self-aware and have onboard artificial intelligence, such that a device at any level can auto-configure itself in accordance with a requirement placed upon it. Field-level devices (IFLDs) can be smart devices which auto-configure based upon a requirement from a higher-level device. Accordingly, system awareness can be incorporated across all levels of the process enabling overall and device-specific optimization of the process. IFLDs can auto-configure to collect and transmit data in accordance with an instruction from a higher-level device, leading to efficient data collection, reduced data bandwidth / processing, and expedited system optimization.

The invention relates to a contrast longitudinal federal map learning method based on adaptive multi-kernel clustering, and the method comprises the following steps: S1, initializing a global model: initializing a global model gG through a server, and distributing the global model gG to all K clients; s2, local training: each client performs local training on the graph learning model by using the graph data held by the client, and model update is generated; and S3, parameter uploading: the K clients send the local graph learning model update theta k to the server. And S4, model aggregation: the server sends the aggregated model update theta global back to the clients, and the server aggregates the model parameter update of the K clients through weighted average to construct a global model. According to the method, multi-source graph knowledge can be integrated in a privacy-protected longitudinal federation environment, and the model is trained under the condition of no label based on a comparative learning mechanism. And self-adaptive multi-core graph clustering is integrated, so that the performance of the model is improved.

The invention belongs to the technical field of federated learning dual optimization, and discloses a federated learning dual optimization method and system based on cluster knowledge distillation and adaptive local differential privacy, and the method comprises the steps: carrying out the data statistical feature extraction of a client; clustering the clients; local training of the client is carried out; applying an adaptive local differential privacy mechanism (ALDP); carrying out model aggregation FedAvg in the cluster; distilling structural knowledge; updating the global model; and evaluating model performance and privacy protection capability. The invention provides a federated learning optimization method which combines client clustering, structural knowledge distillation and a self-adaptive local differential privacy mechanism aiming at the problems of client data heterogeneity, unstable model distillation and difficulty in considering privacy protection and performance in the existing federated learning method. According to the method, client data statistical feature clustering and structural relationship knowledge distillation are combined, and a disturbance mechanism with a dynamic adjustment capability is introduced, so that efficient protection of privacy information is realized while the model performance is improved.

Disclosed in the present application is a blockchain-based distributed ensemble learning method. On the basis of a designed three-layer blockchain structure, which consists of a mini block, an ensemble block and a key block, by means of a consensus protocol, a miner in a network is enabled to train a base model on a training set after performing sampling with replacement, and aggregates models from other miners; and finally, information of the base model and information of an ensemble model are recorded in a blockchain, such that the whole process of model training, model ensemble and model evaluation is fused into a blockchain consensus, and the whole ensemble learning process is automatically executed in a blockchain network. In this way, the present application can increase the utilization rate of the computing power in a blockchain network by means of a proof-of-useful-work mechanism based on machine learning, and a center node is prevented from being introduced during a model aggregation process, thereby maximizing the degree of decentralization of a blockchain.

The invention belongs to the technical field of block chains and federated learning, and particularly discloses a dynamic incentive federated learning method based on a trusted execution environment and a block chain. The method comprises the following steps: firstly, constructing a core computing security area by utilizing TEE, and executing key links such as model aggregation, client screening and contribution measurement in a hardware isolated trusted environment; and then, a block chain and an intelligent contract technology are adopted as a decentralized trust root to realize effective management of identities of participants, tamper-proof records of key certificates and automatic distribution of economic incentives. Secondly, establishing a set of dynamic excitation and reputation mechanism executed in the TEE, and performing credible quantification and automatic reward on the contribution of the client based on multi-dimensional indexes such as model quality, historical reputation, asset pledge, participation stability and the like; and finally, end-to-end data encryption is realized through a session key mechanism in the TEE, so that the data privacy of the client is effectively protected.

The invention discloses a non-independent identically distributed data asynchronous federal learning method based on an improved aggregation algorithm. The method comprises the steps that a server initializes a global model and issues the global model to all clients; and the client performs local training on the received global model by using local data, and uploads the model and model parameters to the server after training is completed. Then, the server adjusts a model lag degree based on a client data volume proportion, calculates model difference consistency, client historical contribution stability, old degree penalty of the client model and cosine similarity of the client model and the global model based on parameters of the client model and the current global model, and generates an asynchronous federal aggregation factor accordingly; and updating the global model parameters to generate a new global model. And finally, testing the global model by the server, and judging whether the learning process is stopped or not. According to the method, fair and effective model aggregation can be realized, and the model convergence stability and the final model detection precision are improved.

The invention provides an intelligent photocuring 3D printing model optimization method, and relates to the technical field of photocuring 3D printing, and the method comprises the following steps: obtaining a 3D printing model file uploaded by a user; analyzing geometric information in the model file, and extracting a triangular grid structure, normal vector information and a surface area; key feature areas of curved surfaces, corners and edges in the model are identified, printing parameter data are collected, and a target model summary file is generated; an optimization printing target is defined, an optimization model is generated based on the generative adversarial network, model quality is evaluated through a discriminator network, a multi-target optimization result is analyzed, and an optimal balance point among different targets is found; and testing the optimized model and the printing parameters by using virtual reality, and performing self-adaptive correction according to a simulation test result to adjust to the optimal printing parameters. According to the method, the geometric features and the material attributes of the 3D model are automatically analyzed, potential weaknesses in the model are identified, and the 3D printing model is intelligently optimized.

The invention discloses an MEC federated learning model aggregation optimization method based on non-independent identically distributed data, and relates to the technical field of edge computing, and the method comprises the following steps: a distributed mobile terminal collects local data, records performance parameters, carries out power law distribution classification after preprocessing, sets a dynamic threshold value, and optimizes local training by using a dynamic sampling algorithm; the asynchronous transmission global model is fused with weight optimization updating and timestamp monitoring technologies to provide heterogeneous network state feedback, and efficient federated learning model aggregation is achieved. According to the method, the efficiency and precision of MEC federated learning under non-independent identically distributed data are remarkably improved, resource allocation is effectively balanced through dynamic threshold adjustment and sampling optimization, the communication overhead is reduced, the model generalization ability is enhanced through asynchronous model updating and weight fusion strategies, real-time feedback of the network state is ensured through the timestamp monitoring technology, and the method has the advantages of being high in reliability and high in reliability. And an efficient and reliable solution is provided for federated learning in the field of edge computing.

The invention provides a multi-task federated learning client contribution degree measurement method based on model layering, and relates to a federated learning and machine learning model layering technology. According to the method, a distributed architecture of a central server and a plurality of client nodes is included, firstly, a federation learning environment and data preparation are configured, then, the central server measures the contribution degree of each client under an existing federation task, and then, the central server layers received model parameters and calculates the gradient of each layer of a model; the model layering similarity relative to the initiator client is calculated to obtain a normalized contribution vector, a new model is obtained through calculation according to the contribution vector and the data size on the client, the client participating in training in the next round is selected, and the task queue is updated. According to the method, the contribution of each client of federal learning can be accurately identified, more accurate model aggregation under layered contribution measurement is realized, meanwhile, the privacy protection effect the same as that of general federal learning is kept, the federal learning time is optimized, and the model precision is improved.

The invention discloses a violation snapshot method based on image recognition, which relates to the related technical field of traffic video monitoring and comprises the steps of environment perception, scene analysis, scene feature library establishment, multi-modal data acquisition, server model aggregation, model issuing and fine adjustment and violation behavior recognition. The invention further discloses a violation snapshot system based on image recognition. The violation snapshot system comprises an environment sensing module, a data acquisition module, a data processing center, a violation behavior recognition module and a snapshot and recording module. Different from a fixed snapshot strategy of a traditional method, the system can dynamically adjust the snapshot strategy according to real-time understanding of a traffic scene, the system can automatically improve the snapshot frame rate when detecting a high-risk scene in which a traffic accident is about to occur, and the snapshot frame rate can be automatically increased in a road section in which the traffic flow is small and the scene is simple. Snapshot resource consumption can be properly reduced; therefore, a complex and changeable traffic environment can be better dealt with, and the snapshot effectiveness and the resource utilization efficiency are improved.

The invention provides a personalized federal learning method suitable for an edge network, which is characterized in that a collaborative training network topology between clients is constructed based on a graph, and local models of the clients and aggregation relationships thereof are described as points and edges of the graph. Secondly, a personalized model aggregation mode based on group dynamics is established, a personalized aggregation matrix is obtained according to the adjacency matrix and the aggregation weight matrix, and finally, the personalized aggregation matrix is used for weighting personalized model parameters uploaded by all clients to obtain a personalized aggregation model; the adjacency matrix is used for controlling the personalized model of the client to move towards the personalized models of other clients aggregated by the client in the state space; the aggregation weight matrix is used for controlling the moving degree of the personalized model; therefore, the personalized model aggregation is rapidly realized in the edge server, and the personalized model automatically completes clustering in the convergence movement in the state space along with the continuous proceeding of training and aggregation, so that the collaborative training of the personalized model between clients is realized.

The invention relates to a privacy protection heterogeneous federal learning method based on split learning. According to the method, federated learning and split learning are combined, privacy protection on client data in distributed collaborative learning is enhanced by adopting a proper model decomposition strategy, and negative effects of data isomerism are reduced by combining a private group normalization layer into a convolutional neural network. The method comprises the following stages: an initialization stage, a model broadcasting stage, a model training updating stage, a model aggregation stage and a training stop judging stage. Compared with the prior art, the method has the advantages that high model prediction accuracy can be kept under the scene of high client data isomerism in the Internet of Things, the data leakage risk is reduced, and the calculation and communication cost of the client is reduced.

The embodiment of the invention discloses a poisoning attack defense method and device based on federated learning, and the method comprises the steps: a central server randomly selects a client participating in training from all clients participating in federated learning, and synchronizes a global model with the client participating in training, and serves as an initial global model; the clients participating in training perform local training on the initial global model to obtain an uploading gradient of a normal client; the central server performs feature extraction on a long-term historical gradient to obtain a score index sequence so as to remove malicious clients in the clients participating in training, and candidate clients are obtained; and the central server performs model aggregation based on the candidate clients, the clients participating in training and the uploading gradient, and updates model parameters of the global model, the updated global model is used as a new round of initial global model, iteration is performed until a preset number of iterations is reached, and the initial global model is obtained. And obtaining a trained global model.

The invention discloses an underwater edge privacy protection task unloading method based on differential federated learning. Firstly, sensor nodes are arranged in a target water area, a network architecture is constructed, a state space, an action space and a reward function are defined respectively, and the sensor nodes execute local model training after sensing environment data. Then, each sensor node uploads local model parameters to a sea surface buoy for model aggregation, and differential disturbance is introduced in the uploading process to protect privacy; and after model aggregation is completed, the sea surface buoy issues a global model to each sensor node, and the node generates a task unloading strategy according to the global model. And finally, the sensor node encrypts the task data through homomorphic encryption, and safely unloads the encrypted calculation task to the corresponding sea surface buoy or low orbit satellite for execution. According to the method, privacy leakage in an underwater task unloading process is effectively reduced, and the security of data transmission in an underwater network is improved.

The invention relates to the technical field of the Internet of Things, in particular to a lightweight distributed model aggregation method applied to a low-power-consumption chip of the Internet of Things, and is suitable for an Internet of Things system composed of a cloud server and a plurality of edge nodes. Each edge node learns an optimal strategy from offline data, and meanwhile, a complex teacher model is compressed into a lightweight student model through a knowledge distillation technology, so that the local calculation burden is reduced. And each edge node uploads quantized and sparse model parameters to the cloud server, and the cloud server performs federal aggregation by adopting a weighted average method to generate a global model and feeds back the global model to each terminal on the premise of ensuring data security. In order to adapt to the practical application of a low-power-consumption chip, a dynamic voltage frequency regulation technology is introduced in the hardware level, and the energy consumption and the communication overhead are effectively reduced. According to the method, the model training process of the edge nodes of the Internet of Things is optimized, and federated learning of privacy protection, efficient communication and cooperative self-adaption is realized.

The invention discloses a federal self-supervised learning backdoor attack method and system based on illusion positive example entanglement, and the method comprises the steps: setting a target of an attacker to embed a backdoor into an image encoder during a pre-training period, and guaranteeing that a downstream classifier trained on the image encoder predicts an image with a trigger as a target class; the limited poisoning data is fully utilized, and the feature similarity of the same trigger sample is improved through the illusion enhancement module, so that the malicious client can effectively embed the backdoor into the global encoder. In the training process, a forged positive sample is generated, learning of the model in a feature space is misguided intentionally, poisoning data is utilized to the maximum extent, an entanglement loss function is designed to optimize a target, the attack efficiency is kept in the whole model aggregation process, and the attenuation risk is reduced. According to the invention, a new perspective and important challenge can be provided for the security research of the federated learning system, and the improvement and development of defense capability are promoted.

The invention discloses a heterogeneous data financial intelligent model collaborative construction device and method and related equipment, and relates to the technical field of model processing, and an institution grouping device of a public control unit clusters and groups financial institutions participating in model construction by using clustering, data distribution speculation and other related technologies. The mechanisms of the same type in the grouping result have similar data distribution characteristics, and the mechanisms in the same group have data with complementary distribution characteristics, that is, missing data knowledge is mutually compensated, and finally good training data with the same distribution are formed on the whole. The grouping result is input into a first model scheduler to schedule mechanism selection of local training of the federal model, the model is guided to complete training and learn complete data knowledge in different grouped mechanisms, and a financial intelligent model meeting result expectation is output through multiple rounds of local model training, correct model scheduling and model aggregation. The method solves the influence of heterogeneous data, and constructs a financial intelligent model with high accuracy.

The present invention belongs to the technical field of communication networks. Disclosed is an adaptive split federated learning method for a distributed AI training service in a 6G computing power network. The method comprises: establishing a 6G computing power network model, and constructing an adaptive split federated learning model for a distributed AI training service; analyzing the delays of each intelligent terminal in a model training stage and a model aggregation stage; constructing a total service delay minimization optimization problem; and using a shortest-path search algorithm to perform joint optimization on model segmentation modes, cooperative node selection modes and multi-domain resource scheduling modes in the model training stage and the model aggregation stage. In the present invention, by means of joint optimization of a model segmentation mode, node selection and multi-dimensional resource allocation in a multi-base-station scenario, the problem of network congestion caused by a single-base-station learning framework processing the aggregation of a large number of intelligent terminal models is alleviated, and the present invention adapts to the heterogeneous computing power of intelligent terminals and the differentiated channel quality in an actual network, thereby improving the efficiency and generalization capability of model training.

The invention provides a decentralized hierarchical federated learning method and system, and edge servers. The method comprises the following steps: a plurality of edge servers respectively receive gradient ciphertexts sent by different participant clients; the plurality of edge servers aggregate the received gradient ciphertext based on a secure multi-party computing protocol to obtain an aggregation model ciphertext; according to the secure multi-party computing protocol, in a non-trusted multi-user network, a plurality of edge servers respectively holding different gradient ciphertexts jointly compute aggregation model ciphertexts based on the gradient ciphertexts, and each edge server only obtains part of data in the aggregation model ciphertexts. The gradient ciphertext held by the edge server is not leaked to other edge servers; and the edge server issues the aggregation model ciphertext to the participant client. By using the scheme of the invention, the security of model aggregation can be improved.

The invention relates to the technical field of network security, in particular to an encrypted network abnormal traffic detection method, system and device and a medium. According to the invention, by combining edge calculation and federated learning, the efficiency and precision of encrypted network traffic anomaly detection are improved. The edge device independently trains a local model and uploads the local model to the central server, and the central server aggregates model parameters through federal learning to generate a global detection model. The method can be closer to a data source for real-time analysis, and delay and bandwidth consumption caused by data transmission are reduced. Meanwhile, by means of local training and model aggregation, concentrated storage and transmission of sensitive data are avoided, and data privacy is effectively guaranteed. The global model is optimized for multiple rounds, abnormal behaviors in encrypted traffic can be accurately identified, the network security protection capability is improved, and an efficient and accurate network security solution with privacy protection is provided.