14 results about "Cognitive technology" patented technology

The invention relates to the technical field of situation cognition, in particular to a maritime multi-mode situation cognition method and device, and the method comprises the steps: training a multi-mode large model, and predicting a maritime situation through employing the trained multi-mode large model; acquiring image data and text data in the same time period; the image encoder performs feature extraction on the image data to obtain image features, and the text encoder performs feature extraction on the text data to obtain text features; and the multi-modal fusion device performs pairing processing on the image features and the text features to obtain cross-modal features, and analyzes the marine situation according to the cross-modal features. Compared with an existing multi-mode large model of a double-tower structure which only performs interaction and fusion of image features and text features in the last layer, the model structure and the feature fusion method adopted by the invention can more fully capture a deep relationship between the image features and the text features; and the fusion effect of the image features and the text features is better.

The invention belongs to the technical field of situation cognition, and particularly discloses an interpretable target group behavior discrimination method, which comprises the following steps: acquiring multi-source situation information data of a detection target group; based on the multi-source situation information data, obtaining a spatial-temporal distribution feature vector of the detection target; constructing a to-be-learned double-layer graph structure based on the multi-source situation information data and the spatial-temporal distribution feature vector; on the basis of the spatial and temporal distribution feature vectors and the to-be-learned graph structure, the behavior category of the detection target group is judged through a relational graph neural network; and through a deep structure causal model SCM, the discrimination result of the relational graph neural network can be explained.

The invention discloses an ultra-high performance concrete knowledge graph construction and mechanism interpretation method based on graph retrieval enhancement generation, and belongs to the technical field of ultra-high performance concrete knowledge graph construction and mechanism cognition. The method comprises the following steps: acquiring a plurality of literatures associated with ultra-high performance concrete; based on a plurality of literatures associated with the ultra-high performance concrete, constructing an ultra-high performance concrete knowledge graph; based on the ultra-high performance concrete knowledge graph, executing self-adaptive length path retrieval on the target origin-destination node to obtain at least one first candidate evidence chain corresponding to the target origin-destination node; and inputting the target instruction and each first candidate evidence chain into the generative model to obtain a corresponding mechanism interpretation text. According to the method, the mechanism explanation of the knowledge of the ultrahigh-performance concrete with high credibility can be efficiently obtained.

The invention relates to the technical field of individualized cognition, in particular to an individualized cognitive decline risk prediction method and system based on multi-mode data fusion. According to the method, multi-modal longitudinal evidence-based data is integrated on the basis of a group-level pathology law, a multi-scale causal link map capable of reflecting the complete period of cognitive decline is formed, pathology time axis alignment and individualized feature normalization are achieved, the dynamic pathology process of individual specificity can be captured, and the accuracy of the pathology process is improved. By fitting individual exclusive multi-scale causal parameters and constructing a time sequence cascade digital twinborn body, forward causal simulation can be executed on multiple time scales, a cognitive decline core driving path is directly revealed, and individual risk quantification and intervention targets are generated. Detection of early hidden pathological signals, dynamic modeling of individual heterogeneity and comprehensive analysis of a multi-channel pathological mechanism are achieved, the ultra-early prediction precision of cognitive decline and individualized intervention suitability are remarkably improved, and a full-period prediction system capable of being iteratively optimized continuously is formed.

The invention relates to an entertainment deductive experience method based on mobile live-action cognition, which belongs to the technical field of mobile live-action cognition and comprises the following steps: detecting a road live-action image and transmitting the road live-action image to an AR information display system for processing; performing efficacy determination and response refresh processing of the Bluetooth network transmission channel; performing control, management and control analysis on the transmission of the road real scene image; the management and control module is used for managing and controlling refreshing of the intelligent camera of the Bluetooth network; according to the method, the transmission efficiency and accuracy of the road live-action image can be efficiently ensured, the complete synchronism of the road live-action image is ensured to the greatest extent, and the road live-action image transmission can be executed by preferably selecting a channel with better efficacy measurement, so that the road live-action image transmission efficacy is still ensured under the condition of a poor Bluetooth network, and the transmission efficiency of the road live-action image is improved. The restoration condition of the Bluetooth network transmission channel can be refreshed when reasonable road live-action image transmission is not suitable, the application efficiency of the Bluetooth network transmission channel is efficiently improved, and the transmission efficiency of the road live-action image is partially improved.

The application discloses a kind of phobia treatment systems based on immersive VR and gaze adaptive virtual cognitions, at least including scene configuration module, task control module, eye movement data acquisition device, data processing module and psychological intervention module, it is applied to the treatment and training in the aspects such as psychological disorder, social skill based on the virtual cognitions (Virtual Cognitions, VCs) technology of gaze point adaptation, especially for the exposure treatment of phobia patient, the system can be applied in large scale, it is favorable to with more efficient, convenient way, improve the treatment effect of phobia and other psychological disorder diseases, help patient to walk out the gloom of fear.

The present application belongs to the technical field of machine high-order cognition, and particularly relates to a method for enhancing the humor reply capability of a pre-training language model of machine high-order cognition.The method comprises the following steps: establishing a connection between a regular text and a humorous text, including constructing an interpretable data set composed of a plurality of preceding text-humor reply pairs, each preceding text-humor reply pair having a humor chain and a humor mind map, and showing the knowledge and logical reasoning process required for generating a humor reply; evaluating and improving the humor reply capability of the pre-training language model, including designing a humor sentiment style classification task and a humor rewriting task as auxiliary humor replies in a multi-task training manner; and designing an encoder-decoder framework, injecting the humor chain and the humor mind map into the pre-training language model, and enhancing the humor reply capability of the pre-training language model based on a multi-task learning manner and using the two auxiliary tasks.The present application can greatly enhance the humor reply capability of the pre-training language model.

The present invention discloses a target functional group identification and a model training method thereof, and belongs to the technical field of situational awareness. The target functional group identification model training method first inputs the normalized data, text data and image data related to the query target into a multimodal large model, and extracts the multimodal feature coding vector of the query target; secondly, the feature coding is passed through a multimodal coding alignment module based on a cross-attention mechanism to achieve vector alignment; then, the aligned feature coding vector is spliced ​​with the query coding vector and input into a multilayer perceptron, and after a SoftMax operation, the probability that the query target belongs to each given functional group is obtained; the model uses a cross entropy loss function to update the parameters of the functional group identification model. Under strong confrontation conditions, the target functional group identification model trained by the present invention can more effectively identify the target functional group.

The application belongs to the technical field of situation awareness, and specifically discloses an interpretable target group behavior discrimination method, which comprises the following steps: acquiring multi-source situation information data of a detected target group; obtaining a space-time distribution feature vector of the detected target based on the multi-source situation information data; constructing a to-be-learned double-layer graph structure based on the multi-source situation information data and the space-time distribution feature vector; discriminating the behavior category of the detected target group through a relational graph neural network based on the space-time distribution feature vector and the to-be-learned graph structure; and making the discrimination result of the relational graph neural network interpretable through a deep structure causal model SCM.

The invention relates to an artificial intelligence cognitive awareness framework technology model, and belongs to the technical field of artificial intelligence technology. The technical problem that the artificial intelligence technology at the present stage is weak in cognitive aspect is mainly solved, and the advanced cognitive ability of artificial intelligence is achieved. According to the method, an artificial intelligence cognitive technology model comprising a nine-layer structure is constructed by simulating a cognitive structure and an intelligence structure of a person, and an artificial intelligence standard knowledge database is constructed, so that logic cognition of artificial intelligence is realized, and cognitive understanding of propositions of a natural form and a social form is realized. Obtaining of artificial intelligence pure knowledge is achieved through knowledge learning cognition, principle cognition of propositions is achieved through thought cognition, obtaining and understanding of artificial intelligence individual experience are achieved through experience cognition, and output of artificial intelligence original achievements is achieved through self-creative cognition. By constructing an artificial intelligence cognition technical framework, professional artificial intelligence cognition application can be provided for various industries.

Present disclosure discloses method and system for automatically generating virtual driving environment using real-world data for autonomous vehicle (AV). Method extracts object feature data and road feature data from information related to an environment surrounding the AV. Thereafter, method generates a dynamic tree-based model for one or more objects in the environment with reference to the AV based on the object feature data and road network information based on the road feature data. Subsequently, method generates a pre-virtual driving environment by combining the dynamic tree-based model for the one or more objects and the road network information using a tree traversal technique and validates the pre-virtual driving environment for inaccuracies based on configurable validation rules. Lastly, method corrects the information based on the inaccuracies obtained during the validation using a cognitive technique and re-generates the virtual driving environment for the AV using the pre-virtual driving environment and the corrected information.

The application relates to the field of intelligent cognitive technology and discloses an improved intelligent cognitive robot cognitive module based on ACT-R, which comprises cognitive modeling intelligent unit modules, feature processing intelligent unit modules, understanding calculation intelligent unit modules, reasoning prediction intelligent unit modules and association recommendation intelligent unit modules which are connected in sequence. The application solves the problems that the existing cognitive calculation does not involve human emotions and implicit knowledge and cannot more truly simulate the human cognitive decision-making process.

The present invention belongs to the field of robot cognitive technology, and specifically is a method for detecting task relationships between objects based on deep learning. The task relationship detection model used includes a pre-trained feature extraction network, a RPN network, an EMA attention module, a three-feature fusion module, and an inference module; the pre-trained feature extraction network extracts features from the input image, the RPN network generates a candidate frame of the object, and a shared subgraph is obtained from the candidate frame; the candidate frame of the object and the shared subgraph are respectively subjected to a region of interest pooling operation to obtain object features and subgraph features; the two object features and one subgraph feature corresponding to the task relationship are respectively subjected to feature extraction by the EMA attention module to obtain subject object features, relationship features, and object object features; the subject object features, relationship features, and object object features are input into the three-feature fusion module for fusion to obtain fused relationship features; the subject object features, object object features, and fused relationship features are subjected to task relationship reasoning by the inference module to obtain a predicted task relationship; the subject and object object features are subjected to category reasoning by the inference module to obtain the predicted category of the object. This method can effectively detect potential task relationships between objects and is used to guide the robot's task operations.

A method for mobile live-action cognition of Chinese calligraphy culture belongs to the technical field of mobile live-action cognition, and combines a suspicion component and tracking software to improve the credibility of confirmation of a tracking information account and guarantee the stability of a penmanship practicing work in a transmission period; the tracking information is grouped, and the reason that the defect is suspicious under abnormal tracking is analyzed and confirmed through the previous tracking task of the corresponding information item of each group of tracking information, so that a worker is cooperated to handle the abnormal task, and a processing thought is given; and moreover, the disposal efficiency is improved, and the complex operation of abnormal retrieval after transfer of the calligraphy practicing works is improved.