101results about How to "Improve learning efficiency" patented technology

The invention discloses a wafer defect detection method and system based on an attention guidance network, and belongs to the technical field of wafer defect detection, and the method comprises the following steps: 1, adjusting light supplement according to a detection demand, and then collecting a wafer image; 2, performing spectrum enhancement preprocessing on the acquired wafer image; and 3, extracting the features of the enhanced image through a multi-scale distributed feature extraction backbone network, and fusing the extracted features through an attention-guided feature pyramid network. Complex background interference of the wafer is effectively suppressed, and defect characteristics are remarkably enhanced; constructing a feature extraction mechanism capable of capturing local details and long-range context information at the same time; balanced and accurate detection of multi-scale defects is realized; the physical priori of the defect is embedded into the network in a learnable manner, so that the learning efficiency and generalization are improved; while ultrahigh precision is ensured, low model complexity is maintained, and the real-time requirement of a production line is met.

The embodiment of the invention provides an online reinforcement learning method and system, a storage medium and a degree product, and the method comprises the steps: controlling a robot to execute a target operation task through a strategy model, so as to obtain corresponding first training data; receiving first interaction information from a human expert to construct corresponding second training data; performing time difference loss calculation based on the first training data and the second training data to train a value evaluation model, and obtaining value evaluation information output by the value evaluation model; performing advantage calculation based on the second training data and the corresponding value evaluation information to obtain advantage information; performing fusion loss calculation based on the advantage information to train a strategy model; the fusion loss comprises advantage weighting strategy loss and near-end strategy optimization loss. According to the embodiment of the invention, the success rate of the robot on the target operation task can be improved under the conditions of lower human teaching cost and fewer environment interaction samples.

The application relates to the technical field of sports training and discloses an intelligent sensing and feedback personalized sports training guidance system, which comprises a dynamic interactive field sensing module, a data processing and analysis module, an immersive feedback module and a personalized model evolution module; the sensing module collects real-time interactive data through an intelligent training pad integrated with pressure and vibration sensors; the data processing and analysis module is internally provided with a predictive intention engine, which can evaluate a posture before a user action and trigger a feedforward early warning, and calculate an actual motion track of the user after the action, namely a fact path; the personalized model evolution module maintains a baseline model based on historical best performance of the user, and the data processing module calls the model to generate an idealized counterfactual path; the immersive feedback module presents an early warning signal on the training pad and synchronously visualizes playback of the fact path and the counterfactual path. The application realizes high self-adaptation and personalization of training guidance and significantly improves training efficiency.

A simulation method and related equipment for machining process training are disclosed. The method includes: acquiring engineering data and process data of the target workpiece; extracting features from the engineering data and combining them with the process data to construct a case resource library; performing semantic association analysis between the pre-set knowledge resource library and the machining cases in the case resource library to generate a process knowledge network; constructing a virtual training environment based on the process knowledge network; collecting interaction data of learners in the virtual training environment and evaluating the interaction data to obtain evaluation results; and generating guidance animations based on the evaluation results and the process knowledge network. This application can improve the learning efficiency of simulation training.

The application discloses a depression electroencephalogram prediction model construction method and device, a storage medium and a computer program product, relates to the technical field of biomedical signal processing and machine learning, and comprises the following steps: extracting a stable signal segment from an original electroencephalogram signal, performing filtering processing on the stable signal segment, and generating an effective signal segment; extracting each feature sequence from the effective signal segment, performing significance screening on each feature sequence, and generating a target electroencephalogram signal feature set; generating a training sample set based on the target electroencephalogram signal feature set, and constructing a probability prediction model through the training sample set. The application solves the technical problem that the model detection effect is poor in the current depression auxiliary diagnosis technology.

The invention relates to the technical field of video processing, and discloses a micro-lesson video processing method and device and a storage medium, in the method, in the process that a recording user records a micro-lesson video through a recording device, after a video clip generating knowledge points is recorded, media supplementary enhancement information of the knowledge points is inserted. In the process that the playing user plays the micro-lesson video through the media player, after the video clips of the knowledge points are played, the interaction content can be displayed to the playing user based on the media supplementation enhancement information of the knowledge points, so that the playing user performs interactive learning based on the interaction content to deeply understand and master the knowledge points in the micro-lesson video. Therefore, when the user performs online learning through the micro-course video, the learning efficiency is improved.

The utility model provides a kind of tightening force monitoring handle and head frame installation force monitoring device, the tightening force monitoring handle can monitor tightening force.Based on the head frame installation force monitoring device of tightening force monitoring handle can be applied in ray operation head frame installation, so that doctor can see the tightening force of left and right hands at any time and adjust at any time, so as to avoid the harm of force overshoot to patient, solve the problem that different position screw tightening force is different, left and right hand tightening force is different, different doctor's tightening force is different and so on due to the screw tightening force has no specific numerical reference.

This application discloses a method and apparatus for pre-training large-scale models based on Bayesian domain reweighting. The method includes: dividing the pre-training corpus into multiple data domains and obtaining corresponding empirical loss signals; learning gamma prior hyperparameters based on the empirical loss signals using a prior prediction network to determine the morphological constraints of the variational posterior distribution; optimizing the morphologically constrained variational posterior distribution based on the optimization objective to obtain a converged domain weight distribution, from which the final domain weights are determined, and using this to perform weighted sampling of the pre-training corpus, achieving mixed sampling of pre-training data for large language models. This solves the problem in existing large-scale model pre-training processes where the learning trajectory of domain weights is unstable, making it impossible to stably and accurately learn the optimal domain weights while simultaneously achieving efficient computation, resulting in low training efficiency and weak generalization ability of large models. This method achieves efficient utilization of multi-domain data and improves the comprehensive performance of large models on different tasks.

The application discloses a kind of reinforcement learning batch process control methods based on improved AC algorithm, it is related to the field of deep reinforcement learning and batch process control field.The method will be based on reinforcement learning method The batch process control is modeled as an optimal control problem on the basis of Markov decision process;Control action constraint is introduced in the reward function of reinforcement learning controller, the number of effective reward samples is increased to improve the learning rate of reinforcement learning controller, and the control cycle is shortened.Priority sampling method is introduced in the Actor-Critic algorithm of deep reinforcement learning, and a soft actor-critic algorithm with priority sampling is proposed to improve the sampling efficiency in the experience replay pool.The present application does not depend on prior knowledge and process model, and can realize model-free control of batch process.

The present application relates to a kind of 5G / 5G-A network slice resource allocation methods supporting high reliability low latency service and enhanced mobile bandwidth service coexistence, belong to mobile communication field.It includes: obtaining URLLC and eMBB service coexistence 5G / 5G-A network parameter information;Optimization target and constraint condition of minimum physical resource block usage are constructed;According to system model and optimization target, establish the hierarchical PPO model assisted by transfer learning, and design state space, action space, reward function and action space truncation mechanism;Pre-training and the parameters of the pre-trained resource allocation model are migrated to hierarchical PPO model;Hierarchical PPO model is trained, and network parameters are updated according to loss function until model converges, obtain the physical resource block allocation method for URLLC and eMBB service coexistence.The present application minimizes physical resource block usage while guaranteeing slice service quality, improves resource utilization.

The specification provides a language model training method, device and program product. The training process of the language model is divided into multiple learning stages. The language model training method comprises: generating multiple constraint conditions for original question samples; based on the original question samples and the corresponding multiple constraint conditions, constructing multiple constraint question samples containing different numbers of constraint conditions; in each learning stage, generating a training set based on all the constructed constraint question samples; each learning stage corresponds to a target number, and the target number increases accordingly as the learning stage progresses; in the training set, the proportion of constraint question samples with the target number of constraint conditions is higher than that of other constraint question samples; training the current language model of each learning stage based on the training set of the learning stage; wherein, from the second learning stage, the current language model of any learning stage is the language model trained in the previous learning stage.

The invention discloses an interactive intelligent teaching system based on multiple modes, and relates to the technical field of intelligent teaching management systems. The system comprises a multi-modal perception layer, a semantic and feature fusion engine, a cognition-emotion two-dimensional analysis module, a strategy decision center and a multi-dimensional interaction execution terminal. Multi-source data of vision, hearing, physiology, behaviors and the like are synchronously collected in real time, and an improved MLP regression network is utilized to construct a cognition-emotion two-dimensional state model of a learner; when the state of the student is monitored to be abnormal, the strategy decision center dynamically generates an optimal teaching path in combination with a knowledge graph and a historical successful case, and drives a virtual digital teacher to execute immersive intervention; according to the method, the problems of emotion perception deficiency, feedback lagging and insufficient individuation in traditional online teaching are solved, and intelligent closed loop and adaptive optimization of the teaching process are realized.

The application discloses a dialect self-adaptive learning method and system, comprising a structured dialect domestication interface, a speech intent recognition unit, a grammar framework domestication unit, a cross-dialect analogy reasoning engine, a dialect-scene-customs linkage memory module and a distributed dialect knowledge federation network; the robot can continuously update dialect recognition capability through daily interaction without intervention of manufacturers. Users can teach the robot to learn dialects in a natural way such as demonstration, error correction and praise, just like teaching children to speak. The system can continuously optimize the model according to the feedback of users, and adapt to the language habits and dialect variants of different users.

A dynamic statistics and result analysis system for watershed river and lake shoreline development and utilization comprises a remote sensing data acquisition module, a data preprocessing and classification module, a database construction module, a change trend analysis module, a river and lake directory and association retrieval module and an interaction management module. The data preprocessing and classification module is used for data preprocessing and statistical classification, the database construction unit is used for constructing a system database, the variation trend analysis module is used for variation trend analysis and statistical report generation, and the river and lake directory and association retrieval module is used for directory maintenance and result association. According to the dynamic statistics and result analysis system for watershed river and lake shoreline development and utilization, a remote sensing data statistical classification algorithm based on a neural network is proposed to classify remote sensing data; and a river-lake shoreline development and utilization change trend analysis algorithm based on deep learning is proposed to analyze the change trend.

The application belongs to the technical field of integrated circuit design, and particularly relates to a cache access system supporting out-of-order processor data prefetching. The system specifically comprises a LOAD memory access information tracking and sequencing module, a LOAD memory access address history buffer, a prefetcher and a target prefetch address buffer. The LOAD memory access information tracking and sequencing module changes out-of-order LOAD memory access information into in-order LOAD memory access information, which is then input into the prefetcher; the prefetcher uses the in-order memory access information to realize more accurate training and target prefetch address prediction; valid target prefetch addresses output by the prefetcher are stored in the target prefetch address buffer to wait for subsequent sending; and the target prefetch address buffer is updated in real time to invalidate untimely addresses, so as to avoid sending useless prefetch addresses. The application can improve the learning efficiency of memory access rules and the accuracy of address prediction, and reduce the resource occupation of prefetch requests on the cache system.

This invention relates to a power data sparsity compression observation method for non-intrusive load monitoring, comprising the following steps: Step 1, acquiring power monitoring data and determining its electricity consumption behavior pattern; Step 2, initializing wavelet sparse basis for sample data based on simple power behavior patterns; Step 3, based on the determination results of the electricity consumption behavior pattern of the completed training samples and the results after wavelet basis initialization in Steps 1 and 2, performing improved K-SVD training and generating sparse basis; Step 4, training the sample dataset Y based on Step 3 to obtain a sparse dictionary D, and performing power monitoring data compression observation based on the improved sparse basis.

The application discloses a data asset standardization packaging method and system and a storage medium, and belongs to the technical field of data processing. The method comprises the following steps: constructing a high-dimensional topological graph based on a business data domain, determining a core asset node set and an external associated node set in the high-dimensional topological graph; calculating the semantic gravity probability of each edge node in the core asset node set to each associated node in the external associated node set based on a semantic gravity field model, and marking the associated node with a semantic gravity probability higher than a preset threshold as a strong gravity boundary node; aggregating the weighted semantic features of all the strong gravity boundary nodes belonging to the same edge node to generate residual semantic metadata; and splicing and encrypting the residual semantic metadata and the naked data ontology of the core asset node set to generate a standard data asset package. The rationality of data asset packaging can be improved by the application.

The invention relates to a language learning system, which comprises a learning system, a video player and a foreign language video material, and is characterized in that the output end of the learning system is electrically connected with the video player, and the foreign language video material is in data connection with the learning system and the video player through a data memory. By adopting the translation algorithm module and the subtitle display module, an English learning program is highly simplified, English learning becomes closer to a native language learning process, information of a foreign language video material is decomposed, extracted, translated and combined, a learner can comprehensively and accurately learn the foreign language video material, and meanwhile, the learning efficiency is improved. The translated subtitle information is transmitted to the video player in real time to be displayed, so that a learner can view the translated subtitle information while watching the video, and the learning efficiency and effect are improved.

The application discloses a large language model test learning method, device, equipment and medium, test data is input into a large language model with low rank adaptive parameters, the prediction score of each token in the vocabulary at each generation step is obtained; the token-level evidence quality of the generation step is extracted based on the prediction score, and the token-level cognitive uncertainty of the generation step is calculated; the token-level cognitive uncertainty is smoothed, the stable area of the generation step is determined according to the smoothed uncertainty, and the mask of the model update window is constructed; the mask entropy loss is calculated based on the mask, and the low rank adaptive parameters are updated through back propagation. The application can solve the problem of adaptive drift in model updating in related technologies.

The application provides a fuel cell unmanned aerial vehicle energy management system, method and unmanned aerial vehicle. The management system comprises a multi-source data acquisition module, an information fusion and state construction module, an integrated prediction and uncertainty quantification module, a risk-sensitive reinforcement learning decision module, a power distribution execution module and an online learning update module. The method of the application constructs a complete technical closed loop of "multi-source information fusion-integrated prediction and uncertainty quantification-risk-sensitive reinforcement learning decision-online evolutionary learning", realizes intelligent balance and unification between real-time performance, global economy and operation safety of energy management in a complex flight environment by constructing an enhanced state space that fuses multi-source real-time information, and combining an integrated prediction model with uncertainty quantification capability and a risk-sensitive deep reinforcement learning decision maker.

This invention provides a method for optimizing BUCK circuit control based on adaptive reward weighting (DQN), comprising the following steps: Step 1, constructing a reinforcement learning control environment for the BUCK circuit; Step 2, designing a reward function based on output voltage error; Step 3, optimizing the reward weight parameters in the reward function using a particle swarm optimization (PSO) algorithm to obtain the optimal reward weight parameters and applying them to the reward function; Step 4, introducing external disturbances into the BUCK circuit control environment and training the DQN agent using the optimized reward function. This invention does not require a precise mathematical model and can achieve adaptive optimization control of the BUCK circuit under different operating conditions, demonstrating significant engineering application value.

This invention relates to the field of immersive educational interactive solutions designing using virtual reality technology, specifically to an immersive educational interactive method and system integrating virtual reality technology. The method includes: data acquisition: collecting students' learning behavior data through virtual reality devices; learning profile generation: generating students' learning profiles based on the learning behavior data using artificial intelligence algorithms; and dynamic adjustment of learning paths: dynamically adjusting the learning path according to the learning profile. The technical solution of this invention, by integrating virtual reality technology and artificial intelligence algorithms, achieves real-time monitoring of students' learning behavior data, dynamic adjustment of learning paths, and generation of personalized learning feedback, significantly improving learning effectiveness, learning interest, and learning persistence. This invention has a higher level of intelligence, stronger adaptability, and more significant educational value, and can provide innovative solutions for personalized learning in the field of education.

The application discloses a quad-rotor formation obstacle avoidance control method based on an improved DDPG algorithm, and belongs to the technical field of unmanned aerial vehicle formation control. The method adopts an improved DDPG reinforcement learning algorithm to plan an obstacle avoidance path of the quad-rotor. When the improved DDPG reinforcement learning algorithm is executed, the priority weight of each quadruple experience in the experience replay pool is initialized. After the quadruple experience in the experience replay pool is sampled and trained, the priority weight of each quadruple experience is recalculated based on a TD error, and a Sum_Tree structure is updated. The method effectively alleviates the training instability caused by hyperparameter sensitivity, the misleading of policy updating caused by overestimation of Q values, and the problem that key experiences are not sufficiently learned, accelerates the convergence speed of the quad-rotor formation obstacle avoidance training, and improves the obstacle avoidance effect.

The invention discloses a nuclear emergency action decision optimization method, system and device and a medium, relates to the technical field of intelligent decision of nuclear accident emergency response and radiation protection, and solves the technical problems that in existing nuclear emergency actions, a multi-source scene changes rapidly, the road network capacity is limited, and real-time risks are difficult to restrain. According to the technical scheme, the method is characterized in that a complex strategy is divided into a high-level strategy and a low-level strategy through a hierarchical reinforcement learning structure, the high-level strategy selects a task target and issues the task target to the low-level strategy, and after the low-level strategy is executed, a result is fed back in a multi-target reward vector form. The high-level strategy adjusts the decision tendency of the high-level strategy according to the accumulated feedback signal, so that a closed-loop control structure of high-level strategy target selection-low-level strategy action execution-high-level strategy target adjustment is formed, task target selection has self-adaptability and upper and lower layer consistency instead of static task allocation, and the task selection efficiency is improved. Therefore, the option selection strategy is optimized step by step, and balance between real-time response and task coherence is considered.

The invention discloses an unmanned aerial vehicle technical data construction method and device, equipment and a storage medium, and relates to the technical field of computers, and the method comprises the steps: extracting target cognitive behavior data of unmanned aerial vehicle aviation support personnel in a task execution process, so as to construct a target triple, and building a cognitive process model through employing a target BERT model; clustering knowledge elements used and maintained by the unmanned aerial vehicle to obtain a knowledge hierarchical framework, extracting unstructured data in the field of the unmanned aerial vehicle, and constructing a target knowledge model by using a knowledge graph technology; determining a cognitive vector of each stage in the cognitive process model; and determining a matching relationship between the cognitive process and the knowledge based on the similarity between the knowledge vector corresponding to the target knowledge model and the cognitive vector, and constructing technical data of the unmanned aerial vehicle in combination with the cognitive process model and the target knowledge model. Cognitive characteristics and knowledge requirements of a user in different stages are considered while the integrity of unmanned aerial vehicle technical data construction is ensured.

The application relates to a cloud service workload prediction method and system based on a convolution enhanced Transformer, which comprises the following steps: collecting cloud server workload data and preprocessing the cloud server workload data to decompose the cloud server workload data into a trend component and a residual component; constructing a workload prediction model, training the workload prediction model by using the trend component and the residual component, and obtaining a trained workload prediction model; predicting the cloud server workload by using the trained workload prediction model to obtain a workload prediction value; predicting the trend component by using a trend information capturing module to obtain a prediction result of the trend component; predicting the residual component by using a convolution enhanced Transformer encoder module to obtain a prediction result of the residual component; and fusing the prediction result of the trend component and the prediction result of the residual component by using a feature fusion module to obtain a final workload prediction value.

The invention discloses a self-organizing learning path optimization method and system based on learning behavior mining, and relates to the field of path optimizing.The self-organizing learning path optimization method based on learning behavior mining comprises the following steps that S1, data are collected, and a library is built; s2, presetting a judgment benchmark, dividing the data to remove redundancy, and integrating to form a data set; s3, preprocessing the data set, extracting core features by using an algorithm through differentiation, and constructing a double-dynamic model in combination with a library; s4, formulating a path criterion, and optimizing to obtain a multi-version push reference and path adaptation range; s5, calling the path candidate set, and generating a personalized self-organizing learning path candidate scheme; s6, implementing the candidate scheme, and collecting data in real time and integrating to form double reports; and S7, a threshold value is preset to judge double reports. According to the invention, through full-period multi-dimensional learning data acquisition and hierarchical cognitive archive library construction, targeted extraction of learning adaptation rhythm and scene adaptation habits is realized.

The invention provides an exercise pushing method and system, equipment and a medium. The pushing method comprises the following steps: acquiring answer information of a user; determining the type of the target exercise according to the answer information; and adjusting the pushing frequency and / or pushing sequence of pushing the target exercises to the user according to the category of the target exercises. According to the exercise pushing method and system, the equipment and the medium, the category of the target exercise is determined by analyzing the answer information of the user, and the pushing frequency and / or the pushing sequence for pushing the target exercise to the user are / is adjusted according to the category of the target exercise, so that a differentiated correction pushing strategy is implemented by accurately identifying wrong question conditions of students with different score levels; students can be helped to accurately make up knowledge vulnerabilities, invalid repeated exercises are reduced, and the learning efficiency of the students is improved.

This invention relates to the field of social media content detection, and more specifically, to an early rumor detection method based on cross-modal meta-learning of social big data with small sample sizes. The method includes: acquiring multimodal information to be detected published on social media; extracting features using a multimodal feature extraction network; constructing a multimodal hidden information extraction network to extract hidden information, thereby obtaining multimodal hidden information; constructing a multimodal fusion processing network; using the multimodal fusion processing network to detect rumors; and optimizing the multimodal hidden information extraction network and the multimodal fusion processing network using a meta-learning algorithm to obtain the final detection result. This invention achieves rumor detection by extracting multimodal hidden information and performing deep fusion processing. Furthermore, this invention achieves early rumor detection with a small sample size by applying a meta-learning algorithm to the multimodal hidden information extraction network and the multimodal fusion processing network.

The application discloses a kind of intelligent recommendation learning method and related device based on reinforcement learning, it is related to education informatization and artificial intelligence technical field, which comprises obtaining student learning data;Initialize student historical answer record and state-action value function table;Based on student historical answer record and hidden Markov model, get continuous mastery state, get discrete mastery state after discretization;Then adopt action selection strategy to recommend candidate exercises in question set, update exercise recommendation sequence;Then update student historical answer record and continuous mastery state, determine total reward, and update state-action value function table, repeatedly iterate to determine the continuous mastery state of student to current knowledge point and exercise recommendation sequence, until satisfying termination condition, obtain optimal exercise recommendation sequence, the application can improve the accuracy of exercise recommendation, to improve the learning efficiency and knowledge mastery level of student.