23 results about "Human intelligence" patented technology

The present invention belongs to the field of artificial intelligence, machine intelligence and intelligent robots. Provided is a method for using human intelligence to compensate for inherent defects of machine intelligence. By means of manipulating multi-perspective visual representations, the respective worlds perceived and constructed by machine intelligence and human intelligence in human-machine inconsistency, respective event narratives of the machine intelligence and the human intelligence in their respective constructed worlds, and spatial-temporal information in the respective event narratives are identified, and are integrated in a unified manner into a fact chain and a logic chain that reflect the complementary nature of the human intelligence and the machine intelligence under the apparent human-computer inconsistency. The application of machine intelligence in human society is evolving, as a whole, from the current imperfect unstable state to a mature and stable state that meets expectations in various aspects of society; In the mature and stable state, predictions can be performed in advance, adjustment can be performed in real time, refusal can be performed post hoc, facts can be confirmed, logic can be clarified, efficient intervention can be performed, correction can be performed in a timely manner, the truth can be understood, efficient communication can be implemented, group consensus can be achieved, the source can be traced post hoc, the rationality can be self-proved, responsibility can be distinguished, responsibility can be determined at each stage, cross-examination can be performed in court, loss can be calculated, and cases can be accumulated.

The invention discloses a light environment ergonomics visual comfort evaluation method based on human intelligence, an edge computing device and a medium. The light environment visual comfort evaluation method comprises the steps of obtaining light environment data for a light environment scene where a user is located; inputting the light environment data into at least one trained visual prediction model for visual comfort correlation analysis to obtain a visual comfort correlation structure corresponding to the light environment data; determining a target visual prediction model based on the visual comfort correlation structure; the light environment scene where the user is located is analyzed through the target visual prediction model, the visual comfort evaluation result is output, and by applying the method, the reliability, accuracy and effectiveness of light environment ergonomics visual comfort evaluation can be improved.

Augmented human intelligence using artificial intelligence (AI) with a human in the loop is disclosed. The AI is trained with data sets with brain-muscle patterns. Brain and muscle signals are obtained for a subject by EEG and EMG sensors to input into the AI. The AI augments human intelligence by overlapping natural human responses with real-time, goal-directed adjustments to instantaneous brain-muscle reactions. The AI enables the human to repeat such adjustments in order to re-wire the brain to adopt new reactions.

The present application relates to a kind of human intelligence collaborative generation type interaction methods of coal mining scene, belong to coal mine intelligent technology field.The present application includes: through coal mine multi-modal intention decoupling-dynamic Agent evaluation and fusion reasoning engine algorithm, the intention of coal mining process scene instruction is identified, subtask division and dependency analysis, obtain structured semantic intention, executable subtask single and task scheduling result and submit to orchestrator, select coal mine professional big model, external tool integrated Agent and decision model integrated Agent are routed by language model empowerment intention analysis database routing technology and are retrieved, obtain the preliminary search result of each Agent;Determine the reliability score of preliminary search result, generate and visual display final search result.The present application makes search result fully fuse multi-modal data, cooperatively consider the search result of different Agent, search result is richer, more accurate, improves the intelligence of human intelligence interaction.

The present invention relates to a fast and efficient human-computer collaborative program translation method, comprising: 1. When a program translation model translates a source program, a software engineer corrects the output code and feeds back the editing results to the program translation model in the form of a prefix; 2. The subsequent retranslation process is divided into two parts: 1). For the generation of prefixes, the cached attention information is used to remove repeated prefixes in the next round. 2). For the remaining generation parts, at each time step, a two-stage method is proposed to determine whether the position can be spliced by a known suffix. If so, the reasoning is terminated in advance. This interaction will be repeated until the software engineer is satisfied with the result. Through the above method, human intelligence and machine intelligence are effectively combined, the accuracy of the final result of program translation is improved, the workload of software engineers is reduced, and the response delay in the collaborative process caused by the excessive length of the program is greatly reduced.

The invention provides an unmanned aerial vehicle combat intention recognition method based on causal characterization, and aims to solve the technical problems of prediction and decision making under non-independent identically distributed data. And the output of human intelligence and machine intelligence is fused by using the Bayesian neural network, so that the robustness and decision-making ability of the system in a complex environment are improved. The model framework designed by the invention is divided into three levels: the first level is independent processing of machine intelligence and human intelligence, the second level is used for converting features into potential causal variables through CE-VAE, and the third level is used for fusing the potential variables through a Bayesian neural network, generating joint causal representation and outputting a final decision.

This application discloses a method, apparatus, and equipment for adjusting navigation tasks based on human intelligence function allocation, belonging to the field of ship control technology. The method includes: performing cross-band dynamic functional connectivity analysis on physiological signals to extract neural state trajectories; performing task semantic parsing on navigation operation logs to extract behavioral log streams corresponding to the ship's sub-task types; constructing positive and negative sample pairs based on the synchronization state of the neural state trajectories and behavioral log streams; mapping the neural state trajectories, behavioral log streams, and environmental pressure vectors to a maritime context-aware latent space with the optimization objectives of maximizing the mutual information of positive sample pairs, minimizing the distance of negative sample pairs, and maximizing the correlation between negative sample pairs and the environmental pressure vector, thereby obtaining a comprehensive contextualized common representation of the crew; and calculating metacognitive alignment loss based on the comprehensive contextualized common representation to adjust the crew's navigation tasks. This method can improve the contextual sensitivity of task adjustment.

The invention discloses a regression test case generation method based on a code large language model, and belongs to the technical field of computer programs. The method comprises the following steps: a step of automatically generating and evaluating a test case oriented to a C language program, a step of generating a test expectation by dialogue with a code large model and using a voting tool, a step of constructing a customized prompt (prompt) template oriented to a C language program type, and a step of generating a test case oriented to a natural language description test requirement. According to the method and the device, the problems of searching in a huge test case space to obtain a high-quality test case and obtaining a reliable test expectation (Oracle) are solved. By combining human intelligence with the generation capability of the code large model, the accuracy, coverage and effectiveness of the test case can be effectively improved, and the problem that a traditional test case generation tool cannot generate test expectation (Oracle) can be solved, so that the software development process is accelerated, and the quality assurance level of software is improved.

The present disclosure relates to transparent and controllable human-AI interaction via chaining of machine learning language models. The present disclosure provides transparent and controllable human-AI interaction via chaining of machine learning language models. In particular, while existing language models (e.g., so-called “large language models” (LLMs)) demonstrate impressive potential on simple tasks, their breadth of scope, lack of transparency, and insufficient controllability can make them less effective when assisting humans with more complex tasks. In response, the present disclosure introduces the concept of chaining instantiations of machine learning language models (e.g., LLMs) together, where the output of one instantiation becomes the input of the next instantiation, and so on, thereby aggregating gains at each step.

The application provides a UAV combat intention judgment method based on a Bayesian model and human-machine complementarity, and the technical scheme is as follows: a human-machine hybrid intelligent framework combining Bayesian inference and human-machine complementarity theory is constructed, real-time situation information of enemy UAVs is collected and analyzed, situation data of the enemy UAVs are subjected to feature extraction, independent intention judgment is carried out by machine intelligence and human intelligence respectively, and relevant parameters are dynamically obtained through Bayesian inference; a confidence feedback mechanism is used to fuse the prediction results of human and machine by combining a human-machine complementarity coefficient; and through dynamic weight distribution and parameter adjustment, more robust and accurate intention recognition results are generated. The application has the beneficial effect that in a complex and dynamic battlefield environment, the complementary advantages of the intuitive judgment of human intelligence and the high-speed calculation of machine intelligence are effectively fused.

The invention relates to a coal mining autonomous decision-making and human intelligence symbiotic multi-agent system, and belongs to the technical field of coal mine intellectualization. Comprising the steps that an equipment control layer is configured with a reflection type program Agent, a distribution decision-making layer is configured with an inference type model Agent, a collaborative decision-making layer is configured with a planning type organization Agent and a planning type arbitration Agent, a capability evolution layer is configured with a learning type reality ring Agent, a learning type simulation ring Agent and a learning type double-ring Agent, and a human-intelligent interaction layer is configured with a generation type task Agent and a generation type investigation Agent. According to the invention, the spanning of an industrial agent from single-machine automation to system autonomy is realized, an Agent community organization mode covering a full coal mining process is formed, systematic design is carried out on multi-Agent role positioning, hierarchical relationship and cooperation rules, and the requirement of real-time cooperative decision making of Agent communities can be met.

The present disclosure provides transparent and controllable human-intelligent interaction via a chain of machine learning language models. In particular, although existing language models (e.g., so-called "Large Language Models" (LLM)) exhibit impressive potential on simple tasks, their breadth of range, lack of transparency and insufficient controllability can make them less effective when assisting humans in performing more complex tasks. In response, the present disclosure introduces the concept of linking instantiations of machine learning language models (e.g., LLMs) together, where the output of one instantiation becomes the input of the next instantiation, and so on, thereby aggregating gains per step.

A method for generating artificially-intelligent, synthetic responses to a natural language conversational survey is provided. Methods create a synthetic persona that reflects an authentic human. Methods store the synthetic persona as vectors within a vector database. Methods initiate a survey. Methods select the synthetic persona from the vector database. The selection is based on a correspondence between data points input by the researcher and vectors included in the synthetic persona. Methods initiate the survey with the selected synthetic persona as a participant. Methods generate a first question for the survey. Methods augment, at the vector database, the first question with vectors that correspond to data points relevant to the first question. Methods transmit the augmented first question to a large language model. Methods receive a response to the augmented first question. Methods process the response at the survey system. Methods enable the researcher to analyze the survey.

The invention relates to the technical field of cognitive feature mining, and particularly discloses a cognitive feature mining method for enhancing a human-intelligent cooperation mechanism. The method comprises the following steps: collecting physiological and psychological characteristics of a user to construct a cognitive preference database; constructing a symbol sequence through a segmented aggregation approximation dimension reduction and symbol approximation theory; extracting an optimized population based on a multi-strategy enhanced escape algorithm, constructing an elite pool to adjust a cognitive boundary, reversely mapping a feature vector into a symbol sequence and labeling an interaction intention to form a symbol data set; and training an LARA model based on the symbol data set, and obtaining the human-intelligent cooperation mutual reliability and evaluating the human-intelligent cooperation mutual reliability in combination with the model recognition accuracy. According to the method, the comprehensiveness of cognitive features is guaranteed through the multi-modal data, the symbol screening accuracy is improved, and the human-intelligent synergy effect is enhanced.

The application provides an intelligent construction project intelligent level evaluation method based on a human intelligence collaborative ternary theory, comprising: obtaining target project data and traditional equipment intelligent correction parameters of each stage; determining total artificial proportion, traditional equipment proportion, human-machine cooperation proportion and intelligent equipment participation proportion in human-machine cooperation of each stage according to the target project data; according to the human intelligence collaborative ternary theory, taking the above proportions and corresponding traditional equipment intelligent correction parameters as evaluation parameters to obtain intelligent degrees of each stage; and obtaining the intelligent degree of the whole life cycle according to the intelligent degrees of each stage and the pre-determined weights of each stage. Through the implementation of the application, the dynamic proportions of artificial, traditional equipment, human-machine cooperation and intelligent equipment independently completing tasks in the construction process are analyzed from the non-intelligent perspective, the globality and coordination of intelligent implementation are ensured, and scientific support is provided for efficient management and implementation of intelligent construction projects.

The invention relates to the cross technical field of intelligent manufacturing, industrial artificial intelligence, a big data technology and high-end material engineering, in particular to a neodymium iron boron production process management method based on artificial intelligence and a big data technology. According to the method, a'big data platform + AI algorithm engine 'double-wheel-driven intelligent management system is constructed, artificial intelligence technologies such as machine learning, deep learning, knowledge graph and digital twinning are deeply fused with a mass multi-source heterogeneous big data processing technology, and the method runs through the whole process of neodymium iron boron material production. The core comprises the steps of constructing a unified process data lake based on a big data technology, and realizing fusion treatment of multi-source data; establishing a quantitative correlation model of process parameters and product performance based on a machine learning algorithm; digital expression and intelligent reasoning of process knowledge are realized based on the knowledge graph; constructing a virtual process simulation optimization platform based on a digital twinning technology; and finally, a continuous evolution closed loop of data acquisition, feature extraction, model training, intelligent decision making and feedback optimization is formed. The technical problems that in traditional neodymium iron boron production, value mining of mass data is insufficient, process decision depends on artificial experience, quality control lags behind, and the process optimization period is long are effectively solved, and fundamental transformation of the production process from experience driving to data and intelligent double-wheel driving is achieved.

The application relates to a medical diagnosis method and system based on human-computer cooperation, which uses BERT+CRF to identify existing symptoms of a patient in a patient conversation, and uses a DQN-based method to let the machine select the next symptom that needs to be asked and confirmed with the patient. After the machine obtains the patient's symptoms, the machine combines a knowledge base to calculate the probability of the patient suffering from a certain disease and generate an electronic medical record of the patient. Human doctors intervene to combine the machine diagnosis results to give a final diagnosis result. The method effectively combines machine intelligence and human intelligence, replaces human doctors with machines to ask for symptoms, greatly reduces the workload of human doctors, and improves online diagnosis efficiency. Meanwhile, the intervention of human doctors can guarantee the reliability of the machine diagnosis result and improve the patient treatment satisfaction.

The invention provides a virtual reality scene-based personnel ability training evaluation method, an edge computing device and a medium, and relates to the technical fields of human intelligence, reality augmentation and the like. The multi-sensory virtual reality-based training evaluation method comprises the steps of outputting non-visual multi-sensory virtual reality information in a virtual reality scene under the condition that a user interacts with a pre-constructed virtual reality scene to execute a training task, and obtaining behavior data and physiological signals of the user; processing the behavior data and the physiological signal to obtain objective evaluation data of the training task executed by the user; obtaining subjective evaluation data for evaluating the training task by the user; and obtaining an ability training evaluation result of the user based on the objective evaluation data and the subjective evaluation data. Through the method, the training evaluation effect is improved, and the perception ability of the user is improved.

The invention discloses a learning type automatic driving method based on deep fusion of a physical model and human intelligence, and relates to the technical field of automatic driving. The method comprises the following steps: constructing an automatic driving decision model based on risk perception, wherein the automatic driving decision model comprises a physical knowledge model for representing longitudinal and transverse safety constraints of a vehicle, a human driver guidance module and a reinforcement learning agent; designing a driving risk field of surrounding vehicles and roads; generating a mixed enhanced guidance action; outputting a final execution action of the autonomous vehicle; and designing a progressive attenuation intervention mechanism to carry out iterative training on the automatic driving decision model. According to the method, a physical model containing driving knowledge and human guidance are fused, and a hybrid enhancement guidance strategy is dynamically optimized through a task delegation mechanism of driving risk perception; human guidance is preferentially adopted in a low-risk scene, and more dependence on a physical model is realized under a high-risk condition, so that a complex driving environment is adapted, the data quality is improved, and reliable and efficient man-machine cooperation is realized.

The application discloses a special object risk management and control prediction method and system based on BERT, and the method comprises the following steps: collecting characteristic data of special objects in past traffic accident cases, and establishing a special object sample library; building a risk management and control prediction model, and training the model by using the special object sample library; collecting characteristic data of special objects to be monitored, establishing a special object monitoring database and updating it in time; calling the risk management and control prediction model, predicting the control means of the special objects to be monitored, and sending early warning information. The application solves the problems of the lack of risk early warning and prevention ability of public security organs for special objects and the incompleteness and subjectivity of traditional human intelligence by constructing a special object sample library and training a risk management and control prediction model, and giving reasonable control measures or processing suggestions for the risks possibly caused by special objects through comprehensive analysis of various text information of the special objects.

The invention discloses an augmented reality comfort evaluation method based on human intelligence and multi-modal data, an edge computing device and a medium, and belongs to the technical field of augmented reality. Comprising the following steps: collecting eye movement data, behavior data and physiological data of a user in an augmented reality environment; performing feature extraction and feature fusion on the eye movement data, the behavior data and the physiological data to obtain multi-modal fusion features; the multi-modal fusion features are input into a multi-modal fusion model, an output result of the multi-modal fusion model is obtained, and the output result at least comprises the user comfort level; the multi-modal fusion model is obtained by taking eye movement data, behavior data and physiological data of a plurality of users under different tasks in the augmented reality environment as sample data and taking subjective comfort scores of the plurality of users under different tasks as labels of the sample data for training. The association between the internal state and the external performance of the user in the augmented reality environment can be mined, and the accuracy of comfort assessment is improved.

The invention relates to a human-intelligent collaborative generation type interaction method for a coal mining scene, and belongs to the technical field of coal mine intellectualization. Comprising the following steps: performing intention recognition, subtask segmentation and dependency relationship analysis on a coal mining process scene instruction through a coal mine multi-mode intention decoupling-dynamic Agent evaluation and fusion inference engine algorithm to obtain a structured semantic intention, an executable subtask list and a task scheduling result, and submitting the structured semantic intention, the executable subtask list and the task scheduling result to a composer; the method comprises the following steps of: selecting a coal mine professional large model, an external tool integrated Agent and a decision model integrated Agent to retrieve through an intention analysis database routing technology endowed by a language model to obtain a preliminary retrieval result of each Agent; and determining a credibility score of the preliminary retrieval result, and generating and visually displaying a final retrieval result. According to the method, the retrieval results are fully fused with multi-modal data, the retrieval results of different Agents are cooperatively considered, the retrieval results are richer and more accurate, and the intelligence of human-intelligent interaction is improved.

The invention provides a cognitive process visualization method, system and device for human-intelligence interaction and a medium, and the method comprises the steps: obtaining a natural language interaction dialogue with artificial intelligence AI, and carrying out high-dimensional semantic analysis and cognitive relationship analysis, so as to convert the natural language interaction dialogue into a structured reasoning event object; according to the reasoning event object, determining a structured cognitive chain used for revealing the thinking path and the AI effect through multi-dimensional relation analysis and stage self-adaption rules; determining a cognitive input index and a creativity index according to a structured cognitive chain, quantized node semantics, a structure and an interaction mode; and according to the cognitive input index and the creativity index, performing automatic layout and visual coding in combination with a structured cognitive chain, and generating a data-driven multi-level visual interface. According to the invention, data-driven multi-level visualization is realized, so that a teacher can realize visualization, diagnosis and intervention of a learning process in an open learning task supported by AI.