Time-domain astronomical transient source identification method and device based on intelligent agent

Abstract

The present invention provides an agent-based time-domain astronomical transient source identification method and device, relating to the field of agent-based data processing and applicable to large-scale model-based astronomical information processing scenarios. The method comprises: inputting observation data into an agent to obtain transient source candidates; the agent obtaining an identification guidance strategy for the transient source candidate from a preset transient source knowledge base; determining the identification steps based on the identification guidance strategy; and in each step, calling a corresponding identification tool to obtain and process multi-band data for the transient source candidate; the agent analyzing the processing results of each identification tool to obtain an identification conclusion; the identification conclusion includes the transient source candidate type, the reasoning process for determining the transient source candidate type, and the annotated key nodes; the reasoning process includes the characteristic data of each identification step. This method lowers the threshold for transient source analysis, determines the credibility of the identification result based on the reasoning process of each identification step, adjusts erroneous reasoning processes, improves accuracy and credibility, and interacts with the user.

Description

Technical Field

[0001] The present invention relates to the field of intelligent agent data processing and is applicable to astronomical information processing scenarios based on large models. More specifically, it relates to an intelligent agent-based time-domain astronomical transient source authentication method, device, equipment, medium and program product. Background Art

[0002] The Einstein Probe (EP) satellite is a representative time-domain astronomy project in my country. Its primary objective is to conduct rapid time-domain surveys of the soft X-ray sky, aiming to discover and explore various high-energy transient and explosive objects in the universe and monitor X-ray variations from these objects. Transient source identification is a core step in EP data processing and analysis. The EP identification process consists of three steps: detecting transient source signals from EP data; identifying and classifying detected transient source signals based on EP observational data and reference data from various sources; and conducting follow-up observations of high-value transient sources to further characterize the sources based on the observational results.

[0003] In the transient source identification of the EP satellite and its experimental satellite LEIA (Lobster Eye Imager for Astronomy), the EP Science Center has set up two positions: Transient Advocate (TA) and Duty Scientist (DS). In order to complete the transient source identification through TA and DS, TA is responsible for reviewing the X-ray data transmitted by the satellite every day, analyzing the transient source candidates therein, and giving preliminary identification results. Then DS decides on the transient source identification results and arranges follow-up observations for high-value transient sources. The most labor-intensive step in this process is TA's analysis of transient source candidates. For a high-confidence transient source candidate, TA needs to call on different tools and reference data for analysis. Although TDAIC (Time Domain Astronomy Information Center, a set of time-domain data analysis software, has provided an integrated data fusion interactive interface, but this process still takes 10-30 minutes. Given the average daily volume of 50 transient source candidates that require TA inspection on the EP satellite, the EP Science Center needs to recruit at least 50 TAs to perform daily transient source identification work. This requires TAs to spend a lot of energy on these relatively basic tasks, resulting in a reduction in energy invested in solving more important scientific problems and reducing scientific output.

[0004] There are also some automated transient source identification methods to solve the problem of low efficiency of manual analysis of transient sources, such as (1) the automated transient source identifier, which can eliminate false sources caused by instruments and data processing to reduce the number of transient source candidates that need manual inspection; (2) the Asteroid Terrestrial Impact Last Alert System (ATLAS) uses a candidate source detection algorithm to identify true and false sources and classify them, and uses a machine learning algorithm to identify the specific category of transient sources and submit them for manual inspection and confirmation; (3) the Zwicky Transient Facility (ZTF) uses a high-brightness transient target identifier based on a neural network to score transient source candidates, and at the same time identifies new transient sources and automatically submits follow-up observation requests; (4) the transient source search algorithm developed by the Swift Space Telescope can achieve low-latency automatic identification of transient sources, but ultimately manual judgment is required to decide whether to conduct follow-up observations. Although these methods have achieved automation to a certain extent, these automated operations are all discriminative operations, which can often only give a specific result and probability value, but cannot provide clear data evidence and reasoning process. For high-value transient source candidates, manual re-judgment based on relevant data is still required, which cannot fundamentally reduce labor costs and improve scientific output efficiency. Summary of the Invention

[0005] In view of the above problems, the present invention provides an intelligent agent-based time-domain astronomical transient source authentication method, device, equipment, medium and program product that reduces manual secondary judgment of transient source candidates and improves scientific output efficiency.

[0006] According to a first aspect of the present invention, there is provided an agent-based time-domain astronomical transient source identification method, comprising: inputting observation data into an agent to obtain transient source candidates; utilizing the agent to obtain an identification guidance strategy for the transient source candidates from a pre-built transient source knowledge base; utilizing the agent to determine the identification steps according to the identification guidance strategy, and in each step calling a corresponding identification tool to obtain and process multi-band data of the transient source candidates; utilizing the agent to analyze the processing results of each identification tool to obtain an identification conclusion; wherein the identification conclusion includes the type of the transient source candidate, the reasoning process for determining the type of the transient source candidate, and the marked key nodes; and the reasoning process includes the feature data of each identification step.

[0007] According to an embodiment of the present invention, based on the authentication guidance strategy, an intelligent agent is used to determine the authentication steps, and in each step, a corresponding authentication tool is called to obtain and process the multi-band data of the transient source candidate, including: based on the authentication guidance strategy, an intelligent agent is used to decompose the authentication task into multiple subtasks to generate authentication steps; and according to the authentication steps, a corresponding authentication tool is called to obtain and process the multi-band data of the transient source candidate.

[0008] According to an embodiment of the present invention, an intelligent agent is used to analyze the processing results of each authentication tool to obtain an authentication conclusion, including: using an intelligent agent to extract the processing results of each authentication tool respectively, and obtaining feature data of each authentication step based on the processing results; calculating the confidence of each authentication step based on the feature data; determining the type of temporary source candidate in each authentication step based on the confidence of each authentication step; and using an intelligent agent to mark nodes in the authentication step whose feature data does not meet preset conditions as key nodes.

[0009] According to an embodiment of the present invention, the reasoning process is obtained in the following manner:

[0010] According to each authentication step and its confidence, and the type of the transient source candidate in each authentication step, a reasoning process is generated according to a preset reasoning method.

[0011] According to an embodiment of the present invention, the method also includes: constructing a visual interface based on the intelligent agent operation log; displaying the verification conclusion and the verification tool call history through the visual interface; if the user finds abnormal content from the verification conclusion and the verification tool call history, guiding the intelligent agent to perform optimization through natural language instructions or operation demonstrations.

[0012] According to an embodiment of the present invention, the method further includes: optimizing the temporary source authentication operation of the intelligent agent by using a dual-path authentication method, a reinforcement learning method, and an evaluation tool.

[0013] Another aspect of the present invention provides an agent-based time-domain astronomical transient source authentication device, comprising:

[0014] A transient source candidate acquisition module is used to input observation data into the intelligent agent to obtain transient source candidates;

[0015] The authentication guidance strategy acquisition module is used to use the intelligent agent to obtain the authentication guidance strategy of the transient source candidate from the pre-built transient source knowledge base;

[0016] The transient source candidate processing module is used to determine the identification steps using the intelligent agent according to the identification guidance strategy, and call the corresponding identification tool in each step to obtain and process the multi-band data of the transient source candidate;

[0017] The authentication conclusion acquisition module is used to use the intelligent agent to analyze the processing results of each authentication tool and obtain the authentication conclusion; wherein, the authentication conclusion includes the type of transient source candidate, the reasoning process for determining the type of transient source candidate and the marked key nodes; the reasoning process includes the feature data of each authentication step.

[0018] Another aspect of an embodiment of the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the method described above.

[0019] Another aspect of an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor is caused to perform the method described above.

[0020] Another aspect of an embodiment of the present invention provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0021] One or more of the above embodiments have the following beneficial effects: they can at least partially address the problem of the existing transient source authentication process lacking reasoning basis and confidence. Therefore, when an intelligent agent presents an authentication conclusion, it can also present the reasoning process, authentication decision, and corresponding confidence level of each step in the authentication process. This allows researchers to determine whether there are any reasoning errors based on the reasoning process, authentication decision, and corresponding confidence level of each step, and to promptly adjust the erroneous reasoning process to optimize the intelligent agent and improve the authentication accuracy.

[0022] It can also at least partially solve the problem of interaction between intelligent agents and scientific researchers, and thus make it possible to guide intelligent agents to answer scientific researchers' questions and query / process / analyze data through natural language, thereby lowering the threshold for data analysis and allowing scientific researchers to devote more energy to the output of scientific results and improve the rate of scientific production. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0024] Figure 1 Schematically illustrates an exemplary system architecture diagram in which various methods and devices described herein may be implemented according to an embodiment of the present invention;

[0025] Figure 2 A flowchart of an agent-based time-domain astronomical transient source authentication method according to an embodiment of the present invention is schematically shown;

[0026] Figure 3 Schematically shows a functional diagram of an intelligent agent according to an embodiment of the present invention;

[0027] Figure 4 A schematic block diagram of a time-domain astronomical transient source authentication device based on an intelligent agent according to an embodiment of the present invention is shown;

[0028] Figure 5 The block diagram of an electronic device suitable for implementing an agent-based time-domain astronomical transient source authentication method according to an embodiment of the present invention is schematically shown.

[0029] It should be noted that, for the sake of clarity, in the drawings used to describe the embodiments of the present invention, the sizes of the overall / partial structures or overall / partial regions may be enlarged or reduced, that is, these drawings are not drawn according to the actual scale. DETAILED DESCRIPTION

[0030] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concept of the present invention.

[0031] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the presence of the features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0032] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0033] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0034] like Figure 1As shown, the system architecture 100 according to this embodiment may include a terminal device 102, a network 103, and a server 104. The network 103 is used to provide a medium for a communication link between the terminal device 102 and the server 104. The network 103 may include various connection types, such as wired and / or wireless communication links, etc.

[0035] A user can use a terminal device 102 to interact with a server 104 via a network 103. The user can send observation data and / or a problem description 101 through the interactive interface provided by the terminal device 102. The terminal device 102 can then send the observation data and / or problem description 101 to the server 104 via the network 103, causing the server 104 to invoke the large model and output a verification conclusion and reply 105. The server then sends the verification conclusion / reply 105 to the terminal device 102, causing the terminal device 102 to display the verification conclusion / reply 105 to the user.

[0036] Various communication client applications may be installed on the terminal device 102, such as intelligent assistant applications, knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software (for example only). Users can enter observation data and / or problem descriptions 101 in the interactive interface of these client applications, and these client applications will display the generated verification conclusion / response content 105 to the user.

[0037] In one embodiment, the server 104 may use a large model to generate the authentication conclusion / response content 105 to display the authentication conclusion / response content 105 on the terminal device 102 .

[0038] The terminal device 102 can be configured as various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, etc.

[0039] Server 104 can be a server that provides various services, such as a backend management server (for example only) that supports the content viewed by users through the interactive interface of terminal device 102. The backend management server can invoke a large model to verify or respond to received observation data and / or question descriptions, and then provide feedback on the verification conclusion / response to terminal device 102 for display through the interactive interface. Server 104 can also be a cloud server, also known as a cloud computing server or cloud host. This is a host product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosts and VPS services ("Virtual Private Servers" or "VPS"). Server 104 can also be a server for a distributed system or a server integrated with blockchain.

[0040] It should be noted that the agent-based time-domain astronomical transient source authentication method provided in the embodiments of the present invention can generally be executed by the server 104. Accordingly, the agent-based time-domain astronomical transient source authentication device provided in the embodiments of the present invention can also be disposed in the server 104. The agent-based time-domain astronomical transient source authentication method provided in the embodiments of the present invention can also be executed by a server or server cluster that is different from the server 104 and that is capable of communicating with the terminal device 102 and / or the server 104. Accordingly, the agent-based time-domain astronomical transient source authentication device provided in the embodiments of the present invention can also be disposed in a server or server cluster that is different from the server 104 and that is capable of communicating with the terminal device 102 and / or the server 104.

[0041] Alternatively, the agent-based time-domain astronomical transient source authentication method provided in the embodiment of the present invention can also be generally executed by the terminal device 102. Accordingly, the agent-based time-domain astronomical transient source authentication device provided in the embodiment of the present invention can generally be set in the terminal device 102.

[0042] It should be understood that Figure 1 The number of terminal devices and servers in the embodiment is merely illustrative. Any number of terminal devices and servers may be used as required.

[0043] In the technical solution of the present invention, the collection, storage, use, processing, transmission, provision, disclosure and application of user personal information involved comply with the provisions of relevant laws and regulations, take necessary confidentiality measures, and do not violate public order and good morals.

[0044] In the technical solution of the present invention, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0045] In view of this, an embodiment of the present invention provides an agent-based time-domain astronomical transient source authentication method. By constructing a transient source knowledge base, the intelligent agent is enhanced and trained, so that the enhanced and trained intelligent agent can perform transient source authentication and provide authentication conclusions. At the same time, the intelligent agent can also interact with users, answer user questions or reanalyze data according to user requirements. Users can also judge whether the reasoning of the authentication process is incorrect based on the provided authentication evidence, and modify the erroneous reasoning process. At the same time, the intelligent agent is optimized, thereby improving the authentication accuracy and reducing labor costs.

[0046] Figure 2 The present invention is a flowchart of an agent-based method for identifying transient astronomical sources in the time domain according to an embodiment of the present invention.

[0047] like Figure 2 As shown, the method 200 includes:

[0048] In operation S210, observation data is input into an agent to obtain a transient source candidate.

[0049] In some embodiments, intelligent agents include systems or entities that can autonomously perceive their environment, make decisions, and execute actions to complete specific tasks. Large models can provide decision support for intelligent agents, empowering them with reasoning, analysis, and task planning capabilities. Intelligent agents leverage the analytical results of large models to execute or optimize their decision-making processes. Intelligent agents can integrate multiple large models to handle different types of tasks.

[0050] A large model can refer to a deep learning model with large-scale model parameters. Large models typically contain hundreds of millions, tens of billions, hundreds of billions, trillions, or even more than ten trillion model parameters. Large models can include large-scale language models, large visual models, large multimodal models, and so on. The large model involved in the embodiments of the present invention can be a general large model, or it can also be an expert large model obtained by fine-tuning based on needs, and the embodiments of the present invention are not limited to this.

[0051] In some embodiments, after the intelligent agent performs preprocessing operations such as calibration and denoising on the observation data, it uses image difference method to identify new or disappeared sources from the preprocessed observation data, or detects mutation points in the preprocessed observation data. It can also use the pulse search method to detect isolated pulses, and use the sources corresponding to the identified new sources, mutation points or isolated pulses as transient source candidates.

[0052] In operation S220 , an intelligent agent is used to obtain a verification guidance strategy for a transient source candidate from a pre-built transient source knowledge base.

[0053] In some embodiments, the transient source knowledge base includes various materials in the transient source field, transient source authentication processes, existing transient source authentication records, and a set of question-answer pairs; wherein the set of question-answer pairs is constructed based on the various materials in the transient source field, the transient source authentication processes, and existing transient source authentication records;

[0054] Furthermore, various materials in the transient source field can be stored in the transient source knowledge base in the form of vector data; the transient source authentication process, existing transient source authentication records and question-answer pair sets can be stored in the transient source knowledge base in the form of knowledge graphs.

[0055] In some embodiments, the agent is fine-tuned based on the set of question-answer pairs so that the fine-tuned agent is more professional and accurate in answering user questions.

[0056] In some embodiments, a search library is constructed based on various materials in the transient source field, and a search enhancement method is used to enhance the intelligent agent so that the enhanced intelligent agent can query the search library for content related to the transient source candidate. When interacting with the user, the enhanced intelligent agent can respond based on the queried content, thereby improving not only the professionalism of the response but also the accuracy of the authentication.

[0057] For example, after a user inputs observation data or a problem description into the agent, the agent selects transient source candidates based on the observation data, generates vector data based on the transient source candidates, or converts the problem description into vector data, and searches for content related to the transient source candidates or the problem description through the vector index. The vector index can be constructed in the following way:

[0058] Use named entity recognition to extract key terms, key content and conclusions from each document in the search database, analyze meta-information such as the title, author, and publication year of each document, convert the extracted key terms, key content and conclusions from the abstract, and meta-information into vector data, and construct a vector index.

[0059] For example, based on the transient source candidate, the infrared band color index of the transient source candidate can be obtained, and content related to the infrared band color index can be found in the retrieval library through vector indexing. The related content includes the infrared band color indices of different celestial body types. By comparing the infrared band color index of the transient source candidate with the infrared band color indices of different celestial body types found, the celestial body type of the transient source candidate can be preliminarily judged. In the subsequent identification process, the infrared band color indices of different celestial body types found can be used as subsequent identification guidance strategies; for example, the question input by the user is described as "What are the light variation characteristics of cosmic rays?" This question focuses on the light variation characteristics of cosmic rays. By searching the retrieval library for content related to the light variation characteristics of cosmic rays through vector indexing, it may be found that the light variation analysis results of cosmic rays present a single-frame bulge. Then the intelligent body will display the found related content as the reply content to the user, that is, "The light variation analysis results of cosmic rays present a single-frame bulge" is displayed to the user as the reply content.

[0060] In some embodiments, the identification guidance strategy includes the steps to be performed to determine the type of the transient source candidate, the data to be called, and how to analyze the data, wherein the called data may include the astronomical characteristics of existing transient sources; for example, various materials in the transient source field record that steps such as multi-band cross-validation and astronomical source table matching are required to determine the type of the transient source candidate, and the transient source identification process and existing transient source identification records record which data to call and how to analyze the data, for example, it is necessary to use the transient source candidate, multi-band reference data and the spectral data of the existing transient source, astrometric data and other data to further determine the type of the transient source candidate; for another example, the data analysis method recorded in the transient source identification process and existing transient source identification records includes judging whether the transient source candidate has a single-frame bulge through the light curve, and if so, it can be preliminarily judged that the transient source candidate is a cosmic ray.

[0061] In operation S230, according to the authentication guidance strategy, the intelligent agent is used to determine the authentication steps, and in each step, a corresponding authentication tool is called to obtain and process the multi-band data of the transient source candidate.

[0062] In some embodiments, before using an intelligent agent to call a corresponding identification tool to process multi-band data of a transient source candidate, the intelligent agent is trained according to a transient source knowledge base so that the trained intelligent agent can call a corresponding identification tool according to the identification steps and learn how to identify the observed data and determine the type of the transient source candidate;

[0063] Specifically, the agent is trained based on the transient source authentication process. Each step of the transient source authentication process is converted into a regularized description, and the authentication tools required for each step are structured and extracted to obtain structured parameters. The regularized description can be: if the authentication parameters of the current step meet the preset execution conditions, then the next step is executed. The regularized description and structured parameters are used as the first training sample to train the agent.

[0064] Since the temporary source authentication process is in natural language text, it is necessary to convert these natural language texts into authentication tool call commands, and then call the corresponding authentication tools through the API. To implement the authentication tool call, the corpus and call commands of each authentication tool in natural language are collected, and the authentication tool call dataset is obtained and used as the second training sample to train the intelligent agent so that the intelligent agent can call the corresponding authentication tool. Among them, the API is a package of various authentication tools (for example, long-term light curve generation tools, multi-band cross-validation tools, astronomical source table matching tools), multi-band reference data, etc. according to the model context protocol. Therefore, the multi-band reference data can be obtained through the API.

[0065] The observation data in the existing transient source identification records are used as the third training sample, and the existing identification conclusions are used as annotations. The intelligent agent is trained based on the third training sample and the annotations, so that the intelligent agent learns to analyze the observation data and obtain the identification conclusion.

[0066] In operation S240, an intelligent agent is used to analyze the processing results of each authentication tool to obtain an authentication conclusion; wherein the authentication conclusion includes the type of transient source candidate, the reasoning process for determining the type of transient source candidate, and the marked key nodes; the reasoning process includes the feature data of each authentication step.

[0067] In an embodiment of the present invention, the intelligent agent enhanced based on retrieval calls the authentication tool and adopts a multimodal data analysis method to realize the authentication of the temporary source. It can not only obtain the type of the temporary source, but also obtain the reasoning process of the authentication. It is beneficial for researchers to check the authentication process according to the authentication conclusion, correct the authentication conclusion with abnormal content and optimize the intelligent agent based on the correction operation, improve the authentication accuracy and reduce labor costs. The intelligent agent interacts with the user, and the user can guide the intelligent agent to re-analyze the data, which lowers the threshold of data analysis and thus improves scientific output.

[0068] In some embodiments, in operation S230, according to the authentication guidance strategy, the intelligent agent determines the authentication steps, and in each step, a corresponding authentication tool is called to obtain and process multi-band data of the transient source candidate, including:

[0069] According to the authentication guidance strategy, the intelligent agent is used to decompose the authentication task into multiple subtasks and generate authentication steps;

[0070] According to the identification steps, call the corresponding identification tool to obtain and process the multi-band data of the transient source candidate.

[0071] For example, obtaining multi-band reference data from the transient source knowledge base to perform the identification task (determine the type of transient source candidate) requires calling multi-band reference data and existing transient source spectral data, X-ray band images, long-term light curves and other data. It is necessary to perform multi-band cross-validation, light variation analysis, spectrum fitting, follow-up observation and other steps. According to these identification guidance strategies, the intelligent agent splits the identification task into four sub-tasks, and generates identification steps based on these four sub-tasks. The intelligent agent performs identification operations step by step according to the generated identification steps. In the process of performing multi-band cross-validation, light variation analysis, and spectrum fitting, the corresponding identification tools are called to perform the corresponding operations. According to the data analysis method learned during the training process, the processing results of each identification tool, the identification guidance strategy and multi-band reference data and other multi-modal data are comprehensively analyzed to obtain the type of the transient source candidate. Finally, the follow-up observation task is performed to mark the transient source candidates of interest for subsequent observation.

[0072] In some embodiments, in operation S240, the intelligent agent is used to analyze the processing results of each authentication tool to obtain an authentication conclusion, including:

[0073] Use the intelligent agent to extract the processing results of each authentication tool respectively, and obtain the feature data of each authentication step based on the processing results;

[0074] Calculate the confidence level of each authentication step based on the feature data;

[0075] Determining the type of the transient source candidate in each authentication step according to the confidence level of each authentication step;

[0076] The intelligent agent is used to mark the nodes whose feature data do not meet the preset conditions in the authentication step as key nodes.

[0077] For example, the authentication tool called by the intelligent agent in the first authentication step is the long-term light curve generation tool, and the authentication tool called in the second authentication step is the X-ray spectrum analysis tool. The characteristic data output by the long-term light curve generation tool and the X-ray spectrum analysis tool are extracted respectively to obtain the light variation characteristics and spectral characteristics of the transient source candidate. In the first authentication step, it is assumed that the light variation characteristics of the transient source candidate present a single-frame bulge, and the extracted authentication guidance strategy has a light variation characteristic of the transient source that shows that the light variation characteristics of cosmic rays present a single-frame bulge. The single-frame bulge of the light variation characteristics of the transient source candidate can be used as an evidence data. Based on the evidence data, the confidence that the transient source candidate obtained in the first authentication step is cosmic rays is calculated. In this embodiment, the light variation characteristics of the transient source candidate are The analysis results show a single-frame bulge, which is judged to be cosmic rays with a confidence level of 98%. In the second identification step, assuming that the spectrum of the transient source candidate is a power-law spectrum, according to the characteristic data of the multi-band reference data, it can be concluded that the transient source with a power-law spectrum is a gamma-ray burst. The power-law spectrum of the transient source candidate can be used as new evidence data. Based on this new evidence data and the confidence level of the previous step, the confidence level of the second identification step is calculated. If the evidence data of the second identification step is that the spectrum of the transient source candidate is a power-law spectrum, the confidence level that the transient source candidate is a gamma-ray burst is the highest, and the transient source type obtained in this identification step can be considered to be a gamma-ray burst. Finally, combining the conclusions of the first and second identification steps, the type of the transient source is finally obtained.

[0078] During the above verification process, feature extraction is performed on the processing results of the long-term light curve generation tool to obtain the brightness change amplitude. If the brightness change amplitude is very small or almost non-existent, it means that there may be a missing optical counterpart and the brightness change amplitude does not meet the preset conditions (brightness change amplitude threshold). Therefore, the node (missing optical counterpart) is marked as a key node so that the user can guide the intelligent agent to re-analyze the data based on the key node.

[0079] In some embodiments, the confidence level of the authentication step is calculated as follows:

[0080]

[0081] in, Indicates that the current step temporary source candidate is Class Source The probability of Indicates the Evidence data, Indicates the current step The marginal likelihood probability of the evidence data is used as a normalization factor. Indicates that the current step's temporary source candidate is Class Source When The probability of Indicates that the evidence data for the next step is When the temporary source candidate is Class Source The posterior probability (i.e., confidence level) of the evidence data can be the data obtained by feature extraction from the data obtained by the authentication tool;

[0082] When calculating the confidence of the first authentication step, Indicates that the transient source candidate is Class Source The prior probability is related to the observation facilities, such as the observation depth, coverage area, sensitivity of the observation instrument, and the occurrence density of various transient sources in different bands. In the absence of evidence data, As an uninformative prior probability, that is, among all transient source types, the transient source candidate is the Class Source The probability of is considered to be uniformly distributed.

[0083] In some embodiments, Calculated by the following formula:

[0084]

[0085] in, Indicates that the current step temporary source candidate is Class Source The probability of Indicates that the current step's temporary source candidate is Class Source When probability.

[0086] In some embodiments, the characteristic data may also include infrared band color index, X-ray band image characteristics, astrometric characteristics, optical band image characteristics, and single light variation characteristics.

[0087] In some embodiments, in operation S240, the reasoning process is obtained according to the following method:

[0088] According to each authentication step and its confidence, and the type of the transient source candidate in each authentication step, a reasoning process is generated according to a preset reasoning method.

[0089] For example, the first identification step is to use the long-term light curve generation tool to process the transient source candidate. In this step, the confidence that the transient source candidate is a cosmic ray is the highest, so the transient source candidate is considered to be a cosmic ray; the second identification step is to use the X-ray spectrum analysis tool to process the transient source candidate. In this step, the confidence that the transient source candidate is a gamma-ray burst is the highest, so the transient source candidate is considered to be a gamma-ray burst; the third identification step is to use the astrometric tool to process the transient source candidate. In this step, the confidence that the transient source candidate is a high-energy gamma-ray burst is the highest, so the transient source candidate is considered to be a high-energy gamma-ray burst; using inductive reasoning, deductive reasoning and other reasoning methods, an inference process is generated, which can be described as: in the first identification step, feature data is extracted from the processing results of the long-term light curve generation tool, and the feature data is compared with the identification guidance strategy or multi-band reference data. The characteristic data of the cosmic rays in the X-ray spectrum are the same, and the confidence that the transient source candidate is a cosmic ray is the highest, so the transient source candidate is determined to be a cosmic ray; in the second identification step, characteristic data are extracted from the processing results of the X-ray spectrum analysis tool, and the characteristic data are the same as the characteristic data of the gamma-ray burst in the identification guidance strategy or the multi-band reference data, and the confidence that the transient source candidate is a gamma-ray burst is the highest, so the transient source candidate is determined to be a gamma-ray burst; in the third identification step, characteristic data are extracted from the processing results of the astrometric tool, and the characteristic data are the same as the characteristic data of the high-energy gamma-ray burst in the identification guidance strategy or the multi-band reference data, and the confidence that the transient source candidate is a high-energy gamma-ray burst is the highest, so the transient source candidate is determined to be a high-energy gamma-ray burst; combining these three identification steps, it is finally obtained that the type of the transient source candidate is a high-energy gamma-ray burst.

[0090] In some embodiments, the method 200 further includes:

[0091] Build a visual interface based on the agent operation log;

[0092] Display the authentication conclusion and authentication tool call history through a visual interface;

[0093] If the user finds abnormal content in the authentication conclusion and authentication tool call history, the agent will be guided to make optimizations through natural language instructions or operation demonstrations;

[0094] For example, the user guides the intelligent agent to modify improper operations through natural language instructions (e.g., prioritizing the identification of transient source candidates that do not match optical counterparts) or operation demonstrations (e.g., marking abnormal intervals of the light curve in the identification evidence on the intelligent agent display interface). The intelligent agent records the modification operation and converts the modification operation into structured data, and adjusts its own model parameters according to the structured data to achieve optimization; wherein, the structured data includes the modification operation type and the parameters that need to be adjusted to perform the modification operation.

[0095] In some embodiments, the visualization interface also displays the confidence level of each authentication step to the user in the form of a heat map;

[0096] The intelligent agent uses a decision path visualization tool to display the calling logic of the authentication tool in real time through a visual interface to assist users in review and reduce the number of iterations of the intelligent agent during training.

[0097] In some embodiments, the identification tool call history records include the X-ray flux data of the transient source candidate and the fitting parameter adjustment records obtained by the identification tool. By displaying the historical status of the identification tool call to the user, it is possible to judge in real time whether the information between each identification step is coherent, thereby improving the identification accuracy.

[0098] In some embodiments, the method 200 further includes:

[0099] A dual-path authentication method, reinforcement learning, and evaluation tools are used to optimize the agent's temporary source authentication operation.

[0100] Furthermore, the dual-path authentication method means that the intelligent agent also constructs a virtual environment and uses a temporary source authentication algorithm to perform temporary source authentication. The authentication conclusion obtained by this method is compared with the authentication conclusion obtained by method 200 to evaluate the authentication performance of the two methods. For example, according to preset scoring criteria, the scores of the two methods in terms of the confidence of the authentication steps and the authentication conclusions are calculated respectively, and the parts of method 200 with low scores are improved. The dual-path authentication method also verifies the end-to-end reliability of the authentication tool call chain.

[0101] Use evaluation tools to record in real time the resource consumption and processing time, accuracy, efficiency, and adaptability of the intelligent authentication process, and improve any areas where performance is poor.

[0102] Optimizing the intelligent agent based on reinforcement learning is to set the reward function and optimization target, and then select the appropriate reinforcement learning method to optimize the intelligent agent based on the reward function and optimization target until the optimization target is achieved.

[0103] Figure 3The figure schematically shows the function diagram of an intelligent agent according to an embodiment of the present invention.

[0104] like Figure 3 As shown, the intelligent agent 310 is trained, enhanced and fine-tuned according to the constructed transient source knowledge base. When the observation data and / or the problem description is input into the trained, enhanced and fine-tuned intelligent agent, the intelligent agent calls the authentication tool according to the authentication step. According to the processing result of the authentication tool, the intelligent agent performs multimodal data analysis in combination with multi-band parameter data, characteristic data of existing transient sources and other multimodal data to obtain evidence data. The confidence of each authentication step is calculated based on the evidence data. The final transient source candidate type is determined according to the confidence of each authentication step. According to each authentication step and its confidence, and the type of transient source candidate in each authentication step, the reasoning process is obtained. The intelligent agent In the process of transient source authentication, the intelligent agent marks key nodes based on multimodal data such as the authentication tool processing results, multi-band parameter data, and characteristic data of existing transient sources. The final transient source candidate type, reasoning process, and marked key nodes are output as the authentication conclusion through a visual interface. The intelligent agent retrieves content related to the problem description in the transient source knowledge base, generates and outputs the response content. The user judges whether the reasoning process is correct based on the output authentication conclusion and makes modifications to incorrect reasoning processes. The intelligent agent adjusts its own model parameters according to the user's modification operations, thereby realizing intelligent agent optimization and improving authentication accuracy and reliability.

[0105] Based on the above-mentioned time domain astronomical transient source identification method, the present invention also provides a time domain astronomical transient source identification device based on intelligent agent. Figure 4 The device is described in detail.

[0106] Figure 4 The structure block diagram of the time-domain astronomical transient source authentication device based on an intelligent agent according to an embodiment of the present invention is schematically shown.

[0107] like Figure 4 As shown, the apparatus 400 includes a transient source candidate acquisition module 410 , an authentication guidance strategy acquisition module 420 , a transient source candidate processing module 430 , and an authentication conclusion acquisition module 440 .

[0108] The transient source candidate acquisition module 410 is used to input observation data into the agent to acquire transient source candidates. In one embodiment, the transient source candidate acquisition module 410 can be used to perform the operation S210 described above, which will not be repeated here.

[0109] The authentication guidance strategy acquisition module 420 is used to use an agent to acquire the authentication guidance strategy of the transient source candidate from the pre-built transient source knowledge base. In one embodiment, the authentication guidance strategy acquisition module 420 can be used to perform the operation S220 described above, which will not be repeated here.

[0110] Transient source candidate processing module 430 is configured to utilize an agent to determine the authentication steps based on the authentication guidance strategy and, in each step, invoke corresponding authentication tools to acquire and process multi-band data for transient source candidates. In one embodiment, transient source candidate processing module 430 can be configured to perform operation S230 described above and will not be further described here.

[0111] Verification conclusion acquisition module 440 is used to use an intelligent agent to analyze the processing results of each verification tool to obtain a verification conclusion. The verification conclusion includes the type of transient source candidate, the reasoning process for determining the transient source candidate type, and the annotated key nodes. The reasoning process includes feature data for each verification step. In one embodiment, verification conclusion acquisition module 440 can be used to perform operation S240 described above and will not be further described here.

[0112] For the parts not mentioned in the apparatus part, they can be understood with reference to the various embodiments of the above-mentioned method. That is, the apparatus part includes modules for executing the various steps of any one of the method embodiments described above. In addition, the implementation methods, technical problems solved, functions achieved, and technical effects achieved of each module / unit / subunit, etc. in the apparatus part embodiment are respectively the same or similar to the implementation methods, technical problems solved, functions achieved, and technical effects achieved of each corresponding step in the method part embodiment, and will not be repeated here.

[0113] According to an embodiment of the present invention, any multiple modules among the transient source candidate acquisition module 410, the authentication guidance strategy acquisition module 420, the transient source candidate processing module 430, and the authentication conclusion acquisition module 440 may be combined into a single module, or any one of these modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module.

[0114] According to an embodiment of the present invention, at least one of the transient source candidate acquisition module 410, the authentication guidance strategy acquisition module 420, the transient source candidate processing module 430, and the authentication conclusion acquisition module 440 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware through any other reasonable method of circuit integration or packaging, or can be implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, at least one of the transient source candidate acquisition module 410, the authentication guidance strategy acquisition module 420, the transient source candidate processing module 430, and the authentication conclusion acquisition module 440 can be at least partially implemented as a computer program module, which can perform the corresponding function when the computer program module is executed.

[0115] Figure 5 The block diagram of an electronic device suitable for implementing an agent-based time-domain astronomical transient source authentication method according to an embodiment of the present invention is schematically shown.

[0116] like Figure 5 As shown, an electronic device 500 according to an embodiment of the present invention includes a processor 501, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 502 or programs loaded from a storage unit 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or related chipsets and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0117] Various programs and data required for the operation of the electronic device 500 are stored in the RAM 503. The processor 501, ROM 502, and RAM 503 are connected to each other via a bus 504. The processor 501 executes the programs in the ROM 502 and / or RAM 503 to perform various operations according to the method flow of the embodiment of the present invention. It should be noted that the programs may also be stored in one or more memories other than the ROM 502 and RAM 503. The processor 501 may also execute the programs stored in the one or more memories to perform various operations according to the method flow of the embodiment of the present invention.

[0118] According to an embodiment of the present invention, electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to bus 504. Electronic device 500 may also include one or more of the following components connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage section 508 including a hard disk; and a communication section 509 including a network interface card such as a LAN card or modem. Communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. Removable media 511, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 510 as needed, so that computer programs read from the removable media can be installed into storage section 508 as needed.

[0119] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present invention.

[0120] According to an embodiment of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, and may include, for example, but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, a computer-readable storage medium may include the ROM 502 and / or RAM 503 described above, and / or one or more memories other than ROM 502 and RAM 503.

[0121] The embodiments of the present invention further include a computer program product, which includes a computer program containing program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to cause the computer system to implement the method provided by the embodiments of the present invention.

[0122] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when the computer program is executed by the processor 501. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0123] In one embodiment, the computer program may be stored on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may be transmitted and distributed in the form of a signal on a network medium, downloaded and installed via the communication portion 509, and / or installed from a removable medium 511. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0124] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 509 and / or installed from the removable medium 511. When the computer program is executed by the processor 501, the above-described functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the systems, devices, means, modules, units, etc. described above can be implemented by computer program modules.

[0125] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0126] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0127] It will be understood by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, even if such combinations or couplings are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention may be combined and / or coupled in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or couplings fall within the scope of the present invention.

[0128] The above describes embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although each embodiment has been described separately above, this does not mean that the measures in each embodiment cannot be advantageously used in combination. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. An agent-based method for identifying transient astronomical sources in the time domain, characterized in that: The method comprises: Input observation data into the intelligent agent to obtain transient source candidates; Using intelligent agents to obtain the authentication guidance strategy of transient source candidates from a pre-built transient source knowledge base; According to the identification guidance strategy, the intelligent agent is used to determine the identification steps, and in each step, the corresponding identification tool is called to obtain and process the multi-band data of the transient source candidate; Use the intelligent agent to analyze the processing results of each authentication tool and obtain the authentication conclusion, including: An intelligent agent is used to extract the processing results of each authentication tool respectively, and based on the processing results, characteristic data of each authentication step is obtained; the confidence of each authentication step is calculated based on the characteristic data; the type of the transient source candidate in each authentication step is determined based on the confidence of each authentication step; the nodes in the authentication step whose characteristic data do not meet the preset conditions are marked as key nodes by the intelligent agent; wherein the authentication conclusion includes the type of the transient source candidate, the reasoning process for determining the type of the transient source candidate and the marked key nodes; the reasoning process includes the characteristic data of each authentication step.

2. The method according to claim 1, characterized in that The method comprises: determining the identification steps using an intelligent agent according to the identification guidance strategy, and calling a corresponding identification tool in each step to obtain and process the multi-band data of the transient source candidate. According to the authentication guidance strategy, the authentication task is decomposed into multiple subtasks by using an intelligent agent to generate authentication steps; According to the identification steps, the corresponding identification tool is called to obtain and process the multi-band data of the transient source candidate.

3. The method according to claim 1, characterized in that The reasoning process is obtained in the following way: According to each authentication step and its confidence, and the type of the transient source candidate in each authentication step, a reasoning process is generated according to a preset reasoning method.

4. The method according to claim 1, wherein The method further comprises: Build a visual interface based on the agent operation log; Display the verification conclusion and verification tool call history through the visual interface; If the user finds abnormal content from the authentication conclusion and authentication tool call history, the agent is guided to perform optimization through natural language instructions or operation demonstrations.

5. The method according to claim 1, wherein The method further comprises: The dual-path authentication method, reinforcement learning method and evaluation tools are used to optimize the temporary source authentication operation of the intelligent agent.

6. An agent-based time-domain astronomical transient source identification device, characterized in that: The device comprises: A transient source candidate acquisition module is used to input observation data into the intelligent agent to obtain transient source candidates; The authentication guidance strategy acquisition module is used to use the intelligent agent to obtain the authentication guidance strategy of the transient source candidate from the pre-built transient source knowledge base; A transient source candidate processing module is used to determine the identification steps using an intelligent agent according to the identification guidance strategy, and to call corresponding identification tools in each step to obtain and process the multi-band data of the transient source candidate; The authentication conclusion acquisition module is used to use an intelligent agent to analyze the processing results of each authentication tool and obtain an authentication conclusion, including: using an intelligent agent to extract the processing results of each authentication tool respectively, and obtaining the characteristic data of each authentication step based on the processing results; calculating the confidence of each authentication step based on the characteristic data; determining the type of the transient source candidate in each authentication step according to the confidence of each authentication step; using an intelligent agent to mark the nodes in the authentication step whose characteristic data does not meet the preset conditions as key nodes; wherein, the authentication conclusion includes the type of the transient source candidate, the reasoning process for determining the type of the transient source candidate and the marked key nodes; the reasoning process includes the characteristic data of each authentication step.

7. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images