18 results about "Web learning" patented technology

This invention relates to a scalable industrial vision task offloading method and scheduling device based on energy consumption awareness, belonging to the field of task scheduling and resource management technology. To address the problems of current task offloading methods failing to simultaneously consider reliability, latency, and energy consumption, and lacking flexibility in adjusting latency and energy consumption, this invention first introduces energy consumption concern weights to establish a hierarchical transmission and computation energy consumption model, and constructs a utility function that comprehensively considers reliability, latency, and energy consumption. Then, task offloading scheduling is modeled as a Markov decision process using deep reinforcement learning, with the utility function as the immediate reward. Through the interaction between the deep reinforcement learning agent and the environment, a policy network is trained to learn the optimal joint decision on execution location and offloading layer under different channel and computing power conditions, thereby maximizing long-term average utility and satisfying latency constraints. Finally, intelligent scheduling is achieved based on deep reinforcement learning.

The application discloses a kind of based on AI model's place name classification method, system, equipment and storage medium, design surveying and mapping technical field, method includes: convergence multi-source professional surveying and mapping place name data, execute data preprocessing, based on the structured place name text feature vector, generate the semantic vector that can be directly identified by general AI big model, input big model extraction place name deep layer semantic structure feature, while extracting the geographical attribute characteristics corresponding to place name;Structural place name field knowledge system library is built;Introduce the basic reasoning framework built by Transformer big language model and traditional deep network learning technology, adopt knowledge enhancement strategy to optimize model, cross-modal reasoning is carried out by fusing place name semantic structure feature and geographical attribute characteristics, preliminarily determine the category of place name;Adopt prompt engineering driven zero sample and few sample learning mode, generate multiple candidate classification results, output final place name classification result after screening verification.

The application discloses a spectrum graph convolution network heterogeneous graph representation learning method based on path collaborative graph enhancement, relates to the technical field of graph neural networks, and comprises the following steps: obtaining to-be-processed heterogeneous graph data, projecting node features of different types into a unified latent feature space by using type-aware linear transformation; calculating structure prior weights; constructing a path collaborative graph, learning path interaction weights by using a graph attention network; constructing a collaborative polynomial spectrum filter; performing spectrum graph convolution based on path collaborative graph enhancement, introducing a learnable positive definite diagonal matrix into the collaborative polynomial spectrum filter, performing weighting and feature transformation on the filtered node features, and obtaining final node representation. The application can learn the importance of meta-paths in a fine-grained manner and capture semantic interaction between meta-paths, solves the problems that existing spectrum heterogeneous graph convolution cannot distinguish the importance of fine-grained paths and lacks semantic collaboration, effectively improves the classification performance of heterogeneous graph nodes, and has a good application prospect.

1. The name of the design product: wisdom in the screen (TB-CE007). 2. The use of the design product: the product is used for wisdom in the screen, and the main functions are audio and video interaction, Internet television playing, smart home control, home security, office video conference, home network learning, etc. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

A Petri net-based task scheduling method for heterogeneous environments relates to the field of collaborative decision-making and resource optimization in edge computing systems. During scheduling, the input tasks and their dependencies are first modeled as directed acyclic graphs (DAGs), and then converted into Petri nets to uniformly represent task states, resource constraints, and device collaboration relationships, in conjunction with device status information. Subsequently, graph attention networks are used to extract PN topological features, and a dynamic transition triggering strategy is learned using a dual-deep Q-network. The task unloading and resource allocation sequences are optimized through a reward function, and the sequence validity is further verified by a sequence detection module. During task execution, tasks are scheduled in real-time according to a reinforcement learning strategy, and the network state and decision model are dynamically updated based on environmental feedback, forming a closed-loop optimization. This invention significantly improves task throughput and resource utilization in heterogeneous edge environments, and is particularly suitable for device-to-device collaborative scenarios with complex task dependencies and dynamically changing resources.

This invention discloses an information retrieval optimization method based on few-sample knowledge graph completion. It constructs a few-sample knowledge graph completion model, GAKDN, and utilizes a gated and role-aware neighbor aggregator to effectively filter neighbor noise information, reducing the impact of noise on entity embedding learning. A role recognition network learns deep associations between entities and their neighbors, identifying the roles of entities under different few-sample relationships. Knowledge distillation extracts structural information between entity pairs corresponding to few-sample relationships, alleviating the problem of insufficient sample representation. An adaptive matching processor calculates the scores of the positive and negative query sets and the support set, finding the most suitable entity for completion based on the highest score. This invention solves the problems of noisy neighborhoods, multiple entity roles, and insufficient sample feature learning in few-sample scenarios, thereby more accurately predicting the tail entity to be completed, improving the accuracy of knowledge graph completion in few-sample scenarios, and optimizing the information retrieval capability of knowledge graphs.

1. The name of the design product: smart middle screen (TB-CE003). 2. The use of the design product: The product is used for smart middle screen, and the main functions are audio and video interaction, Internet TV playing, smart home control, home security, office video conference, home network learning, etc. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

This invention discloses a microblog topic mining method based on text enhancement and high-order conversational relationship guidance, including: (1) constructing a user-level dialogue network containing high-order conversational relationships; (2) user embedding based on text enhancement: using a text enhancement method based on a large language model to interpret colloquial words to utilize their semantic information, and using a pooling function based on a multilayer perceptron to learn the mutual influence between different words, generating user representations based on text sequences; (3) self-fusion network representation: capturing the nonlinear relationship between text content and high-order network structure, and introducing an attention mechanism to model the influence of different users on the topic in the sequence, obtaining user sequence representations; (4) topic generation based on neural variational reasoning: using the user sequence representation with structural information supplementation as input to neural variational reasoning, learning potential generation factors through network reconstruction, and finally generating topics with better consistency.

The application discloses a kind of based on digital twinning's Internet of Vehicles edge computing unloading multi-objective decision-making method, specifically: input based on digital twinning's edge Internet of Vehicles environment, initialize executor-evaluator network parameters and based on the parameter of multi-objective evolutionary algorithm based on decomposition;Perform unloading action, obtain reward vector and completion flag, store in cache area;According to the data fitting value function of cache area, based on current value function and discount return calculation advantage estimate;Using multi-objective evolutionary algorithm based on decomposition and reward vector update solution set and fitness value, return Pareto optimal solution as executor network learning parameter;With truncation method constraint policy update, calculate loss function and update strategy;When completion flag is true, reset edge Internet of Vehicles environment starts next round.The application is suitable for in unknown dynamic edge Internet of Vehicles environment through digital twinning auxiliary intelligent edge unloading decision to realize the long-term trade-off between minimization delay, energy consumption and cloud computing cost.

The present application relates to the technical field of knowledge distillation, and particularly relates to a cross-modal knowledge distillation method based on non-radar modal to radar modal. The method comprises the following steps: acquiring synchronous historical visual modal data and historical radar modal data, and obtaining time-aligned visual input data and radar input data after preprocessing; constructing a teacher-student knowledge distillation network, taking the visual input data as the input of the teacher network and the radar input data as the input of the student network for training; the teacher network and the student network both adopt a space-time feature decoupling architecture, the teacher network extracts long-term time features of the visual modal, the student network extracts long-term time features of the radar modal, and the student network learns the long-term time features extracted by the teacher network through knowledge distillation; after the training is completed, only the student network is deployed for inference and recognition of the radar modal data. The present application realizes stable knowledge migration from visual to radar, significantly improves the radar recognition accuracy, and meets the edge lightweight deployment requirements.

This invention discloses a heterogeneous attribute-enhanced collaborative comparison product recommendation method based on a large language model, belonging to the field of intelligent recommendation technology. The method first obtains collaborative representations of users and products through a graph neural network based on a user-product interaction graph. Then, it constructs an attribute value relationship graph based on user attribute value data and product attribute value data, and learns attribute value node representations using a graph attention network. On this basis, a large language model is introduced for semantic reasoning to obtain a set of heterogeneous attribute value information related to the semantics of static attribute values. Furthermore, the heterogeneous attribute value information is fused with the static attribute value information, and an attribute-aware representation is obtained through an attention mechanism. This representation is then gated and fused with the collaborative representation to form the final representation. Finally, the model is optimized by combining Bayesian personalized ranking loss and contrastive learning loss, and Top-n recommendation results are generated based on the predicted scores. This invention can improve the accuracy and personalization level of recommendation results.

The application discloses an atomic skill incremental learning method for near-field fine operation and belongs to the technical field of robots. The method comprises the following steps: collecting ontology and visual-tactile data under different near-field tasks, and realizing atomic skill decomposition based on multi-modal feature clustering; constructing a multi-scale space-time feature coding network, learning the dynamic characteristic evolution track of a target object in a near-field operation process; introducing a Hebb-type network local connection update mechanism, constructing an atomic skill memory matrix based on the principal components of the evolution track, and realizing atomic skill category incremental learning; outputting a task-level atomic skill combination sequence according to input near-field multi-modal data, and online optimizing a continuous force-position control track. Through the cooperation of space-time dynamic modeling and Hebb-type local connection update, the application realizes rapid incremental recursion and continuous track generation of atomic skills, and improves the stability, adaptability and generalization ability of robots in complex and limited scenes.

The application discloses a command sequence anomaly detection method and system based on semantic representation and context perception, the method comprises the following steps: constructing a multi-layer command filtering and cleaning mechanism, adopting a three-level processing flow of session reconstruction, command deduplication and noise filtering to convert original commands into standardized token sequences; designing a semantic representation learning module, training an encoder-decoder network through a mask language model to learn deep semantic representation of commands; constructing a context perception modeling module, capturing context dependency of command sequences based on a state space model and a sliding window mechanism; implementing a two-stage anomaly detection strategy; the system comprises a command preprocessing module, a context semantic modeling module and an anomaly detection module. The application realizes fine detection from the session level to the command level, effectively controls the computing overhead while ensuring the detection accuracy, greatly reduces the model construction and maintenance cost, and improves the practicability and deployability of the system.

This invention proposes an incomplete multimodal learning method based on hierarchical prompts and knowledge distillation, belonging to the field of artificial intelligence technology. The method includes: acquiring and preprocessing a multimodal sample dataset; extracting modal features of different modalities using a multimodal encoder to obtain modality alignment features; constructing and calling a modality prompt generator to generate affinity modality-level prompts, and training a teacher network on a small number of complete modality samples; inputting the modality alignment features into an instance prompt generator to generate instance-level prompts, and training a student network on a large number of missing modality samples; and fine-tuning the teacher and student networks based on a triple knowledge distillation mechanism. This invention possesses stronger adaptability and generalization capabilities. By guiding student network learning through the training of the teacher network on complete data, it can better handle modality missing situations; simultaneously, it eliminates the need for separate modeling of each missing combination, significantly improving the system's scalability.

A computer based method for an electronic marketing campaign from a customer to a contact receives a campaign having an email message body and a historic profile of a previous campaign by the customer. The email message body includes text and image data. The email message body is preprocessed based upon the campaign and the historic profile to produce campaign training data. A neural network learning model is trained with the campaign training data. The neural network provides a subject line recommendation inference, and named entity recognition is performed on the subject line recommendation.

A system comprising a computer comprising a processor and a memory. The memory comprises instructions causing the processor to be programmed to: receive an intermediate concept constraint at a neural network; and train the neural network with training data, training labels, and at least one of a data constraint, a feature constraint, or the intermediate concept constraint.

1. The name of the design product: wisdom in the screen (E8-R). 2. The use of the design product: the product is used for wisdom in the screen, and the main functions are audio and video interaction, Internet television playing, smart home control, home security, office video conference, home network learning, etc. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

The application provides an open-vocabulary few-shot three-dimensional part segmentation method and system, and relates to the technical field of image processing. A 2D teacher network is used to process a three-dimensional point cloud sample to generate a 3D knowledge unit containing part semantic information; the three-dimensional point cloud sample and the 3D knowledge unit are input into a 3D student network including an encoder, an adapter and a segmentation head, and training is carried out based on knowledge distillation by using a double-layer circulation mechanism; the double-layer circulation mechanism is respectively used for enabling the network to learn task-specific prior knowledge and cross-class universal prior knowledge; in the reasoning stage, a search enhancement strategy is used to dynamically select a support set for a target point cloud, and a segmentation head is fine-tuned through an inner-layer circulation mechanism to obtain a part segmentation result of the target point cloud. The application effectively bridges the representation difference between a two-dimensional visual language model and three-dimensional point cloud processing, breaks through the limitation of a traditional single-class model, and realizes rapid adaptation to unknown object classes and high-quality open-vocabulary few-shot three-dimensional part segmentation.