9 results about "Boltzmann machine" patented technology

The present application belongs to the technical field of data processing, and discloses a multi-modal project data analysis method, which comprises the following steps: collecting heterogeneous modal project data; based on adaptive multi-scale filtering technology, denoising and timestamp correction are performed on each modal project data, and cross-modal time alignment is performed, and a multi-modal signal is output; a multi-level Boltzmann machine energy network is constructed, and a joint probability distribution of the multi-modal signal is modeled; a dynamic complexity measurement mechanism based on topological entropy is designed, which is used for representing project structure complexity, and a topological entropy dynamic field is generated; the topological entropy dynamic field is used as a constraint condition, a multi-dimensional covariant field framework is used, nonlinear coupling and space-time propagation among the multi-modal signals are simulated, the multi-modal signals are abstracted into string vibration modes, local topological defects in the topological entropy dynamic field are identified, and an abnormal fluctuation atlas is generated; and the analysis of complex multi-modal data in the project running process is more intelligent and controllable.

The application discloses a kind of water transport ship traffic collaborative control methods based on deep learning, specifically includes: obtaining ship operation data set, time index alignment is carried out, and state vector sequence is generated;State vector sequence is encoded to feature, and deep boltzmann machine model is constructed;First conditional energy function, second conditional energy function and energy coupling matrix are established between model layer, and form hierarchical conditional energy function set;Hierarchical conditional energy function set is carried out to hidden variable probability inference, and hidden layer probability state vector sequence is generated;Parameter flow shape is divided to energy function parameter, and policy optimization is carried out under the constraint of confidence domain, and updated energy function parameter set is generated;Updated energy function parameter is carried out to energy calculation, and state probability distribution sequence is generated;State probability distribution sequence is carried out to strategy mapping, and ship traffic collaborative control strategy is generated.The application constructs hierarchical energy coupling model, realizes the collaborative control of multiple ships state.

The application discloses a production business process node vector representation method fusing structure and behavior characteristics. The method comprises the following steps: firstly, structure characteristics of a production business process are extracted by using a random walk algorithm and a Skip-gram model; then, behavior characteristics of the production business process are extracted by using the Skip-gram model; finally, a deep Boltzmann machine is introduced to fuse the structure characteristics and the behavior characteristics of the production business process, thereby automatically performing vector representation on nodes of the production business process, extracting and fusing various characteristics to improve the accuracy of the node vector representation, and further improving the accuracy of subsequent production business process recommendation, optimization and the like. The application has the advantages that the graph structure characteristics of the production business process itself and the behavior characteristics based on historical execution logs are fused, deep learning technology is used to strengthen the production business process model, the node vector representation which simply depends on the historical execution logs is supplemented and improved, and the accuracy of the node vector representation is improved.

The invention discloses a personalized learning path recommendation method based on deep learning, and the method comprises the following steps: S1, collecting and standardizing a learning log, resource data and a knowledge point first repair relation, and generating a training data set and a knowledge graph; s2, constructing a deep Boltzmann machine model based on the training data set and initializing parameters; s3, performing model improvement based on meta learning; s4, training a strategy prior network and a value network; s5, screening resources according to the knowledge graph and the learner state to generate a candidate set; s6, adopting a Monte Carlo tree with prior guidance to search and optimize path selection, and generating candidate learning paths; and S7, selecting an optimal path according to the candidate path scores, and outputting final model parameters after judging convergence. According to the method, accurate recommendation of personalized learning paths is realized, and the learning efficiency and the intelligent adaptability of the model are improved.

The application discloses a kind of multi-modal characteristic fusion's intelligent target identification method, on the basis of constructing typical aerial target high-resolution range image, infrared image sample library, respectively using attention bidirectional gate recurrent unit model from high-resolution range image time series sample, using light convolutional neural network model from infrared image sample learning extraction aerial target multi-modal deep feature representation;Using Gaussian distribution limited Boltzmann machine realizes aerial target multi-modal feature abstract fusion, removes redundancy, forms more discernibility, representative multi-modal feature joint representation;Using kernel ultra-limit learning machine with strong generalization ability under small sample condition, fast training speed and high classification accuracy as classifier, realize aerial target type identification.The application uses heterogeneous deep learning model to abstract extraction and fusion aerial target feature information in high-resolution range image, infrared image for target identification, improve recognition precision.

This invention discloses an AI-based intelligent patrol method and system for unmanned aerial vehicles (UAVs), comprising the following steps: collecting multi-source perception data from the UAV and preprocessing it; performing multi-modal feature fusion and spatial topology modeling using an improved multi-modal deep Boltzmann machine; constructing a three-dimensional coordinate reference system to determine tower locations, conductor directions, ground feature heights, and navigable airspace; performing path search and constraint solving to generate an initial inspection route; performing end-side inference based on a pruned quantization multi-modal deep Boltzmann machine and updating the unified environmental spatial representation; incrementally updating the three-dimensional spatial modeling results and correcting the trajectory to generate an optimized inspection route; collecting multi-source perception data along the optimized inspection route, performing association labeling and data uploading to generate an inspection dataset. This invention achieves intelligent patrol of UAVs based on multi-modal deep learning and three-dimensional spatial modeling, possessing advantages such as accurate environmental perception, adaptive route optimization, and high-quality management of inspection data.

This invention provides an ADAS intelligent control method based on full-vehicle camera perception, comprising the following steps: S1: Rigid body spatiotemporal synchronization and online joint calibration of the full-vehicle surround-view camera group based on Lie group and Lie algebra; S2: Pixel-level alignment and full-domain static-dynamic bi-branch semantic mask generation; S3: Hierarchical confidence perception and target full attribute decoupling extraction based on variational Bayesian inference; S4: Spatiotemporal coupled risk field modeling based on lattice Boltzmann machine; S5: Multi-constraint hierarchical decision-making and expected behavior sequence generation; S6: Actuator control quantity calculation and feedforward compensation. This invention breaks down the independent barriers between each link in traditional schemes by deeply integrating the Lie group and Lie algebra online calibration algorithm, the variational Bayesian decoupled perception algorithm, and the lattice Boltzmann machine risk field modeling algorithm, achieving end-to-end collaborative optimization and closed-loop iteration from front-end calibration to end-to-end control.

The application discloses a kind of based on Boltzmann's model neural network prediction method, comprising: by Boltzmann machine model creation with symmetrical connection right random neural network and deploy log server;With log, the network topology of Boltzmann machine is handled in layers, obtains first matrix and second matrix;First matrix and second matrix are respectively put into Boltzmann machine state transition Markov chain and are predicted, if the difference of predicted value is less than 10% weighted average, then matrix predicted value is final predicted value;If the difference of predicted value is greater than 10% weighted average, then take the final predicted value of the proportion of high core business data of matrix data. The application guarantees the timeliness of data processing, and highlights the position of artificial intelligence in the field of big data processing.

The present invention relates to the technical field of protein structure prediction, and relates to an optical quantum computer-based protein structure prediction method, system and apparatus. The method comprises: 1) obtaining a multiple sequence alignment (MSA) matrix of a target sequence on the basis of target sequence alignment of a protein needing to be predicted; 2) encoding the obtained MSA matrix into a (0,1) matrix; 3) converting interaction of amino acids of the protein into an undirected graph model of (0,1) state nodes, and using a Boltzmann machine training mechanism and an optical quantum computer to perform training to obtain weight coefficients of edges connecting nodes in the undirected graph model; and 4) obtaining coefficients of interaction between different amino acids of the protein on the basis of the weight coefficients. The present invention improves the computation efficiency and solution result of protein structure prediction, and solves the problems in the prior art that it is difficult to perform computation for complex scheduling problems and accurate solutions cannot be obtained.