50 results about "Modularity (networks)" patented technology

The invention provides a power distribution network communication network fault diagnosis and self-adaptive reconstruction method, which comprises the steps of formulating corresponding layering and partitioning strategies aiming at different network scales and service types, including modular management of a large-scale multi-layer network and centralized response of a small-scale network; the faults of various networks can be effectively positioned and isolated; recognizing a reconstruction object after the fault positioning is completed, and formulating a network reconstruction scheme according to the recognized reconstruction object and the current state of the network, including but not limited to replacing a fault node, reconstructing a connection or modifying a route by starting a standby resource; after the reconstruction scheme is executed, the data collection and analysis period is adjusted according to the fault frequency, and the reconstruction interval is controlled to reduce redundant network structure adjustment and potential network fluctuation, including active detection and passive monitoring of the power distribution network and bidirectional monitoring of the network condition.

A scalable platform for orchestrating networks of collaborative AI agents utilizing modular hybrid computing architecture. The platform integrates classical, quantum, and neuromorphic computing paradigms through hardware-accelerated translation layers and cross-paradigm coordination mechanisms. A central orchestration engine manages interactions between domain-specific AI agents, dynamically distributing workloads across heterogeneous computing cores based on task complexity, computational requirements, and resource availability. The platform employs hardware-accelerated translation between paradigms, enabling efficient cross-paradigm information exchange while maintaining semantic consistency and computation integrity across different architectures. Specialized monitoring and optimization systems continuously adjust resource allocation and fine-tune performance across computing paradigms. Advanced cache management and fault tolerance mechanisms ensure reliable operation, while privacy-preservation techniques enable secure collaboration. The platform's modular architecture supports integration of different computational approaches, enabling complex multi-domain problem solving that leverages the unique advantages of each paradigm while maintaining system-wide efficiency, scalability, and coherence.

The invention relates to the technical field of artificial intelligence graph neural networks (GNN), in particular to a graph neural network pre-training method and system based on community structure generality. The pre-training method comprises the following steps: preprocessing an input graph to form a graph G (V, A) comprising a node set V and an edge set A; community structure detection is carried out on the graph G (V, A) to identify the community structure of the graph G (V, A), and a community is highly identified by adopting a modularity-based Louvain method or a label propagation LPA method; the invention provides a graph neural network pre-training method based on community structure generality. The problems that in the prior art, large-scale graph data processing efficiency is low, and dependence on labeled data is high can be effectively solved. Through preprocessing and community structure detection on an input graph, a modularity optimized Louvain method or a label propagation LPA method is adopted, and communities can be efficiently identified and divided, so that local and global structure information of the graph is reserved.

An electronic circuit system implementing and executing machine learning inference engines. While ML inference engines are based on (architectures and parameters defined by) configured, trained and tuned machine learning models, our design has the novel ability to support data driven, on-the-fly-reconfigured model runs. Reconfiguration and tuning operations include dynamic computational graph modifications, define-by-run alterations, changes to network depth (number of layers) and width (neurons per layer), and adjustments to weights, biases, plus activation function parameters. Neural networks supported include Feed-Forward, RNN, CNN, and Hopfield architectures, plus Ensemble, Federated, Cooperating, Adversarial, and Swarm collections. Decision Trees and Forests are also supported, as are more esoteric approaches such as ART and KAN. Our invention is capable of running both standalone and cooperatively, the cooperative processing being local and / or remote / cloud based, interfacing with telemetry applications to feed data, and machine learning software to feed new or updated models.

The current disclosure provides methods and systems that can manage the dynamic-uncertainty induced by resources of decentralized data networks, in order to ensure the stability and sustainability of task-oriented automated operations, such as operations conducted through intelligent agents. In contrast to the state-of-the-art object-based processing, the methods and systems enable object-aware processing, to ensure the establishment of stable and sustainable associations with physical and digital web-objects, while enabling those objects to be processed dynamically with full adaptability in response to contextual and structural alterations in real-time. Thus, the disclosure provides—both human and machine—users with the ability to develop and deploy modular systems capable of engaging any conceivable physical or digital interaction with the resources of a data network, such as dynamically linking and manipulating clusters of complex web-objects to execute complex tasks in complex and dynamic web environments—most importantly—stably and sustainably.

The invention discloses a water supply network optimization scheduling system based on clustering grouping and Cauchy variation, and belongs to the field of water supply network optimization scheduling. Through a multi-dimensional comprehensive optimization algorithm, an intelligent water consumption prediction model, a self-adaptive pressure partitioning technology, an intelligent pump station collaborative scheduling system, a comprehensive cost-benefit analysis tool, real-time data-driven dynamic scheduling, an interdisciplinary technology fusion platform and an environment-friendly scheduling strategy, the intelligent pump station collaborative scheduling system and the comprehensive cost-benefit analysis tool are integrated. The intelligent and automatic management of the water supply pipe network is realized through the innovation points of intelligent control, automatic operation, modularized and extensible software architecture and the like; data are collected in real time by using the Internet of Things technology, high-precision water consumption prediction is performed in combination with a deep learning model, and the water supply efficiency is optimized and the operation cost is reduced by dynamically adjusting pressure partition and pump station collaborative scheduling; green energy is preferentially used, new function integration is supported through modular design and an extensible framework, the system adapts to the requirements of different regions and pipe networks, and the long-term sustainable development capacity of the system is improved.

Multi-stage hybrid network integrates relationship regularization links and explainable elements to improve alignment with human values, explainability, robustness, and efficiency. The network comprises neural components, event prediction elements, and probability models across multiple stages, with relationship constraints enforcing structured knowledge representation. Explainable elements provide interpretable rationales for decisions, enhancing transparency. Training incorporates supervised learning, human-guided refinement, semi-automated knowledge engineering, and adversarial robustness techniques. A Socratic reasoning module detects contradictions and refines outputs for logical consistency. Indexed model elements enable dynamic memory optimization for improved efficiency. Candidate outputs may be scored, verified, or selected using neural and symbolic criteria. The invention supports retry loops and configurable subsystem pipelines to improve output quality. Applications include text generation, speech recognition, translation, and decision support. By combining structured constraints, human oversight, and modular architectures, the system improves the trustworthiness, safety, and adaptability of AI systems across diverse modalities and tasks.

The invention provides a weighting network key node identification method based on layering. The method comprises the following steps: step 1, establishing a multi-layer network initial model based on a layering network thought; 2, selecting two layers of networks with the maximum similar modularity gain after combination to perform iterative combination, and obtaining a multi-layer network model by taking the similar modularity gains of any two layers of networks are non-positive as a loop termination condition; step 3, considering betweenness indexes of each network node in the multi-layer network model, measuring importance of each network node in the multi-layer network model, and dividing the network nodes into interactive nodes, first-level non-interactive nodes and second-level non-interactive nodes; and 4, evaluating the importance of the interactive nodes, the primary non-interactive nodes and the secondary non-interactive nodes by adopting three betweenness indexes. According to the method, multi-layer network modeling is carried out on the premise that the global attributes of the key nodes of the complex network have limitation, and the recognition accuracy of the key nodes in the network nodes of the multi-layer network is ensured.

A system is provided for optimizing dynamic routing in a capsule network. The system includes an autoencoder configured to encode input data into a latent space representation; a plurality of capsule networks comprising capsule layers; and a generative adversarial network (GAN) including (a) a generator that receives the latent space representation and a noise vector and outputs routing coefficients, and (b) a discriminator that evaluates the routing coefficients by assessing performance metrics from one or more of the capsule networks. The routing coefficients determine how outputs from a capsule layer in one capsule network are routed to a capsule layer in another. Based on performance feedback, the discriminator adjusts the generator for subsequent iterations to improve routing effectiveness. This architecture enables adaptive, performance-optimized routing across modular capsule networks using adversarial learning.

The present invention discloses a brain network classification method based on community detection and dual-path autoencoding. We introduce a community detection method to detect the potential community structure in the brain network, and then, according to the result of the community detection, readjust the sorting order of the nodes in the network. On the one hand, it allows the modular structure in the network to be retained, and on the other hand, the adjusted brain network is more suitable for the rectangular convolution kernel in the deep learning method. Since brain network data has two modes, namely structural mode and functional mode. In order to better integrate the information of the two modes, the present invention adopts dual-path autoencoding technology to classify brain networks. First, we use the structural mode to guide the community detection and network structure adjustment method of the functional mode network. Then, the dual-path autoencoding technology with an attention mechanism is used to extract important high-order information in the network. Finally, the complementary information of the structural mode and the functional mode is comprehensively utilized to complete the classification task of the brain network.

The application discloses a drainage pipe network hydraulics experimental method and system based on double-branch dynamic coupling, belongs to the technical field of water conservancy projects, and comprises the following steps: a double-branch double-gradient dynamic coupling simulation device is built, four types of visual simulation experiments such as pipe network intersection flow state analysis are carried out, and data are collected; after the data are preprocessed, a critical blockage flow prediction model is established, a four-dimensional model is constructed in combination with historical data of a city drainage pipe network to train and iterate the four-dimensional model; field data are collected in real time to input the model for prediction, and the pipe network is controlled in real time according to the results and the model is continuously updated. The system comprises experimental simulation, model establishment, model training and comprehensive control modules. The application realizes multi-parameter dynamic collaborative control, constructs a multi-dimensional accurate measurement system, has the advantages of modularity and water cycle, realizes prediction control integration, provides an active prevention and control means for safe operation of the drainage pipe network, and is suitable for visual research on the hydraulic characteristics of the pipe network.

A polarization community detection method based on a multi-layer symbol network is characterized in that a single-layer symbol network is constructed for each community, nodes of the single-layer symbol network represent individuals in the community, positive sides among the nodes represent friendly relationships among the individuals, and negative sides among the nodes represent hostile relationships among the individuals; the interaction relationship among individuals in different communities is represented by cross-layer coupling edges, and the single-layer symbol networks are connected through the cross-layer coupling edges to form a multi-layer symbol network; initializing a multi-layer symbol network, independently dividing each node into a community, and calculating an initial value of the modularity of the multi-layer symbol network; carrying out community division again by taking the initial value of the modularity of the multilayer symbol network as an initial comparison reference, traversing and transferring all nodes in a loop iteration mode until the modularity of the multilayer symbol network reaches the maximum value through community division, completing community division, and obtaining a plurality of new communities as polarized communities; and calculating a polarity value of each new community, and judging the polarized communities based on the polarity values.

This invention discloses a method for reconfigurable modular data plane pipeline assembly and deterministic execution based on XDP. The method includes: generating a Profile configuration file based on the target application scenario, wherein the Profile configuration file describes the module information of the modules to be assembled; obtaining the target information of all modules to be assembled through a loader based on the pre-configured self-describing metadata of the modules, and generating an assembly input list; loading and parsing the Profile configuration file to generate a mapping relationship from stages to module entry points, and writing the mapping relationship into a stage scheduling table; reading the stage scheduling table through the XDP entry scheduler, jumping to the execution entry point of the corresponding module according to the stage scheduling table, and scheduling the execution of the ingress stage chain and the egress stage chain of modules in sequence. This invention abstracts structural changes such as inserting new functional segments, deleting modules, rearranging execution order, and separating ingress and egress processing chains into an assembly problem, thus making it more suitable for building service chains and traffic weaving in physical networks.

An AI-driven network security system that proactively detects and mitigates threats in enterprise environments is provided. The system integrates AI-based behavioral analysis to model normal activity and identify anomalies. It implements a Reverse Authentication Algorithm (RAA) to generate a unique 32-hex-character (128-bit) authentication (“au”) string formed from random hexadecimal values, a complement-masked timestamp, and a search string dynamically derived from an ASCII random string table. Incoming requests are validated by reconstructing and verifying the timestamp and by recomputing the search string extracted from the au string. Upon anomaly detection or authentication failure, the system triggers layered defenses including packet filtering, IP blocking via a blocklist, execution of custom threat-neutralization scripts, and integration with content delivery networks (CDNs) to mitigate volumetric attacks. The modular architecture enables real-time detection and response while integrating with existing infrastructure.

This application provides a method, system, equipment, and medium for the operation and maintenance of equipment in underground engineering networks. By constructing a modular network architecture with one master and multiple slaves and a network operation and maintenance management software system, it achieves unified management, dynamic expansion, and proactive operation and maintenance of heterogeneous equipment in the underground engineering network. Specifically, the method includes: acquiring the equipment information set and network operation dataset of each device in the underground engineering network, wherein the underground engineering network adopts a modular network architecture with one master and multiple slaves; mapping the devices to logical resource entities based on the equipment information set; subsequently, generating a network topology diagram corresponding to the underground engineering network based on each logical resource entity; and finally, generating an equipment operation and maintenance diagram of the underground engineering network based on the network operation dataset and the network topology diagram. This method aims to improve the efficiency of equipment operation and maintenance in underground engineering networks.

This invention relates to the field of intelligent design technology for anchoring systems, specifically to a method and system for predicting the response of anchoring systems based on knowledge distillation and neural networks. The method includes the following steps: S1, acquiring first input data for the anchoring system, the first input data including working condition parameters, material parameters, component selection parameters, and structural geometric parameters; the anchoring system includes a multi-module structure composed of several identical single structural modules; S2, inputting the first input data into a student model for knowledge distillation to obtain the structural response of the anchoring system. The method of this invention directly uses a student model to predict the structural response. Starting from a teacher model trained on a single modular structure, knowledge distillation technology can guide the final student model, reducing the size of the structural dataset required for direct training while simultaneously meeting the ideal prediction accuracy requirements.

The invention discloses a search model establishment method and system based on a heterogeneous graph neural network and application, and belongs to the technical field of artificial intelligence and graph neural networks. The search model establishment method comprises the following steps: collecting meta-paths related to a target node to form an initial meta-path set; generating a meta-path redundancy graph by utilizing the meta-paths of which the dependency degree between every two in the initial meta-path set is higher than a preset threshold value; searching an optimal non-adjacent meta-path combination from the meta-path redundancy graph as an initial non-redundant meta-path combination; according to meta-paths and meta-path strength in the initial non-redundant meta-path combination, optimizing architecture parameters and network weights of the initial non-redundant meta-path combination to obtain an optimal non-redundant meta-path combination; establishing a meta-path aggregation search space as a search model for the optimal non-redundant meta-path combination; and optimizing and training a search model by using a micromodular architecture search strategy. According to the method, the performance and efficiency of representing the learning task by the heterogeneous graph are improved.

This invention presents a method and system for extracting highly timely time-series data from multi-source heterogeneous networks. The system employs asynchronous message queues to achieve loosely coupled distributed collaborative operation of five functional modules: web crawling, structured parsing, data cleaning and mapping, task scheduling, and data storage. By integrating a multi-mode crawling engine with static, dynamic rendering, and API interface calls, the system solves the compatibility problem of heterogeneous data sources. The structured parsing module uses a hybrid parsing approach combining rule bases and a BERT-based natural language processing model, significantly improving the intelligence and accuracy of entity extraction from unstructured text. The data cleaning module uses the statistical Z-Score algorithm to dynamically detect outliers in time-series data, ensuring the reliability of the stored data. The task scheduling module supports incremental updates and failure retry mechanisms, greatly improving system throughput and operational stability. This invention achieves automated, highly reliable, and highly timely time-series data acquisition, cleaning, and structured storage through an advanced modular architecture, AI-driven parsing technology, and statistical data quality control mechanisms.

A fully autonomous and unsupervised system for automated AI red teaming that utilizes generative adversarial networks (GANs), reinforcement learning (RL), and modular compliance logic to evaluate the robustness, reliability, and regulatory compliance of AI systems. The invention provides an adaptive adversarial testing engine that can simulate real-world attacks on AI models, log outcomes, generate audit reports, and assist in aligning model behavior with global AI governance standards such as ISO 42001, NIST AI RMF, and national AI certification frameworks. Designed for real-time and scalable deployment across diverse AI modalities, the system supports integration with vision, language, structured data, and multi-modal AI systems, and is designed for scalable deployment via API infrastructure. The invention is designed for scalable deployment in AI governance environments, including use in certification workflows, regulatory compliance assurance, and infrastructure protection across institutional and enterprise settings.

The present invention provides a method and system for detecting extended communities based on community clustering features, comprising the following steps: based on the clustering features of communities, using a core community extension algorithm to obtain extended communities; using a community optimization algorithm based on modularity to optimize the extended communities, and completing the community detection of complex networks. Aiming at the problem that community clustering features are not considered, the present invention extends communities with community clustering features during the seed extension process; aiming at the problem that sparse community screening is inaccurate during the community optimization process, during community optimization, the present invention introduces modularity increment and merges communities with modularity maximization, further improving the accuracy of community detection.

Systems and methods for providing an artificial intelligence-based solution in a dynamic environment that requires models with varying degrees of nuance and specialization. One such dynamic environment relates to generating dynamic human-like conversational responses based on complex data. In particular, systems and methods recite generating dynamic human-like conversational responses using a modular architecture featuring layered data models with gated neural networks. The modular architecture compartmentalizes the various components and functions of an application. That is, the architecture may support multiple layers, each featuring models performing specific functions and / or having been trained on using specific data and / or algorithms.

The invention provides an artificial-intelligence–based platform for automated analysis of Pap smear slides. The system integrates preprocessing, dual deep learning models (CNN-Transformer and EfficientNet-LSTM), and reporting modules to classify cytological images into diagnostic categories including NILM, ASC-US, LSIL, HSIL, and SCC. The platform includes interpretability outputs, secure data management, and active learning capabilities, enabling scalable, accurate, and transparent cervical-cancer screening across diverse laboratory settings. The platform is suitable for implementation in clinical pathology laboratories, telemedicine networks, and population-level screening programs. It allows rapid, reproducible, and objective evaluation of cytology slides, improving throughput while reducing reliance on manual slide examination. The invention is deployable on local servers, private clouds, or public cloud infrastructures and is adaptable to various cytology specimens. Adoption of this system enhances early detection of precancerous lesions, reduces inter-observer variability, and facilitates integration with laboratory information systems. It also supports continuous learning and adaptation to evolving imaging protocols. The invention provides multiple advantages over existing methods: (1) Dual-model architecture combining CNN-Transformer and EfficientNet- LSTM ensures robust classification across both dense and sparse cellular regions. (2)Preprocessing and adaptive tiling improve feature extraction from variable-quality images. (3)Interpretability modules enhance clinician trust and facilitate regulatory compliance. (4) Active learning allows continuous improvement based on expert feedback. (5) Modular deployment supports both local laboratory and cloud-based operation. (6) Data security features meet international standards for privacy and encryption. (7) High scalability enables screening of large slide volumes without compromising accuracy. The invention provides a clinically relevant, efficient, and technologically advanced solution for automated cervical-cancer screening, applicable in diverse healthcare environments and adaptable to future cytology domains.

This invention discloses a dynamic zoning method for distribution networks based on a multi-objective ant colony algorithm. The method includes: defining a binary modularity using a tracking algorithm to quantify the energy balance within a zoning interval; proposing a multi-objective ant colony algorithm suitable for distribution network zoning by combining a power reserve index; ensuring power balance in each zoning interval and strong control capability of the power source within the zoning interval; combining a probabilistic prediction scenario set with the zoning algorithm to obtain dynamic active power zoning, improving the zoning interval's ability to cope with the uncertainty of renewable energy and enhancing the robustness of the zoning scheme; obtaining a Pareto optimal solution set by solving the final power reserve and final binary modularity of the distribution network zoning using the multi-objective ant colony optimization algorithm; selecting the optimal zoning scheme; realizing dynamic zoning of the distribution network; improving zoning efficiency; and reducing the impact of subjective human selection.

The invention relates to the technical field of production line layout, in particular to a production line layout optimization method based on multi-agent reinforcement learning, and the method comprises the steps: constructing a topological structure of a production line through employing a directed weighted network based on a complex network theory; carrying out community division on nodes in the topological structure based on a modularity optimized condensed hierarchical clustering algorithm to obtain a plurality of communities; and carrying out optimized layout on the equipment in each community based on a multi-agent layout algorithm of MAPPO to complete layout of all the equipment in each community, and then carrying out manual layout on all the communities to obtain an optimized production line layout. According to the invention, on the basis of production line equipment community classification of a complex network, modular recombination of equipment clusters is realized through community division, and the cross-community logistics complexity is reduced.

The application discloses a search model establishment method and system based on a heterogeneous graph neural network and application, and belongs to the technical field of artificial intelligence and graph neural networks. The search model establishment method comprises the following steps: collecting meta-paths related to a target node to form an initial meta-path set; generating a meta-path redundancy graph by using meta-paths in the initial meta-path set, which have a dependency higher than a preset threshold; finding an optimal non-adjacent meta-path combination from the meta-path redundancy graph as an initial non-redundant meta-path combination; optimizing the architecture parameters and network weights of the initial non-redundant meta-path combination according to the meta-paths and meta-path strengths in the initial non-redundant meta-path combination to obtain an optimal non-redundant meta-path combination; establishing a meta-path aggregation search space for the optimal non-redundant meta-path combination as a search model; and optimizing and training the search model by using a differentiable modular architecture search strategy. The application improves the performance and efficiency of a heterogeneous graph representation learning task.

The invention discloses an overlapping community-based rumor suppression maximization method, which comprises the following steps of: firstly, dividing a social network related to a rumor source into a plurality of communities based on a democratic estimation algorithm of a network modular organization, and identifying overlapping communities and overlapping nodes in the communities; secondly, distributing a positive seed node budget for each community in proportion according to the community scale, screening out the positive seed nodes with the most influence in each community according to the trust center value, selecting the positive seed nodes meeting the community positive seed node budget from the positive seed nodes, forming a candidate positive seed node set, and incorporating the candidate positive seed node set into overlapped nodes; and finally, optimizing the candidate positive seed node set by using a genetic algorithm to generate an optimal positive seed node set. According to the method, the network position attributes of the nodes are comprehensively considered, the precision and effect of the positive sub-nodes are improved, good adaptability and expansibility are achieved in networks of different scales, and the dual targets of rumor suppression and true phase propagation are effectively achieved.

This invention discloses a method, apparatus, and medium for multi-scale wavelet analysis of brain networks based on manifold learning. Using T1-weighted MRI and DW-MRI images, combined with Desctrieux mapping and probabilistic fiber tractography based on surface seeds, an initial adjacency matrix is ​​obtained. The average adjacency matrix is ​​calculated. Based on the node degree, betweenness, PageRank, and assignment coefficient of the average adjacency matrix, several nodes are selected from the brain network. Masks at different scales are calculated on the nodes. Multi-scale wavelets are initialized. The eigenvectors of the Laplacian matrix of the average adjacency matrix are solved using a power iteration method to obtain the optimal multi-scale wavelet. Protein signals in the brain network are projected onto this wavelet to obtain new biomarker signals from the brain network. Therefore, this embodiment of the invention uses manifold learning to calculate the mean of a brain network group, which better maintains the geometric topology of the network and, considering the hierarchical modularity and centrality of network nodes, better uncovers some potential physiological and pathological mechanisms in brain diseases.

The present invention discloses a method for optimizing the partitioning of complex electromagnetic ring networks based on the Louvain algorithm. The method comprises simplifying the receiving-end power grid model into a weighted undirected topological graph, partitioning the nodes of the weighted undirected topological graph into multiple communities based on the Louvain algorithm, and assigning each node to the community where its adjacent nodes reside until the modularity of each community reaches its maximum value. When the communities formed after all nodes have been moved and the three-phase short-circuit currents of each partition meet the judgment criteria, a partitioning scheme is output, and all partitioning schemes are evaluated for indicators to select the optimal partitioning scheme. The method provided by the present invention aims to improve the operating efficiency and stability of the power grid while reducing the complexity in power grid planning and operation.

The current disclosure provides methods and systems that can manage the dynamic-uncertainty induced by resources of decentralized data networks, in order to ensure the stability and sustainability of task-oriented automated operations, such as operations conducted through intelligent agents. In contrast to the state-of-the-art object-based processing, the methods and systems enable object-aware processing, to ensure the establishment of stable and sustainable associations with physical and digital web-objects, while enabling those objects to be processed dynamically with full adaptability in response to contextual and structural alterations in real-time. Thus, the disclosure provides—both human and machine—users with the ability to develop and deploy modular systems capable of engaging any conceivable physical or digital interaction with the resources of a data network, such as dynamically linking and manipulating clusters of complex web-objects to execute complex tasks in complex and dynamic web environments—most importantly—stably and sustainably.