最新刊期
卷 6 ， 期 3 ， 2024
- 摘要：The further development of intelligent science and technology will inevitably transform the knowledge infrastructure of society, compelling us to reorganize the K21 education system—from early childhood to doctoral stages—using new thinking and new organizations with new operations. This will, in turn, redefine the functions and responsibilities of future schools, governments, social institutions, and public organizations.关键词：K21 education;smart education;parallel education;intelligent technology;artificial intelligence;parallel intelligence210|1131|2更新时间：2024-09-15
Review Intelligence
- 摘要：In the digital society of "open sharing of information and interconnection of everything", the information traces characterized with "massive, multi-source and explosive growth" in the internet provide rich "mineral deposits" for open-source intelligence collecting. Intelligence processing enabled by advanced artificial intelligence technologies, such as natural language processing and computer vision, greatly improves the efficiency of intelligence production and is widely used in the field of security. However, generative artificial intelligence represented by deepfake has opened the "Pandora's box" of artificial intelligence and is used to create digital wildfires and put false information, which confuse the public and bring crucial challenges to the integration of intelligent intelligence. This paper focuses on open-source intelligence and deeply analyzes its characteristics and existing challenges. Then, this paper summarizes the development of advanced artificial intelligence technology in opensource intelligence fusion, summarizes the existing intelligence spoofing attack methods and intelligence anti-spoofing defense methods from the perspective of confrontation. Finally, new directions of intelligence fusion in the future are proposed to provide reference for the trusted intelligence fusion and provide support for intelligent situation analysis and auxiliary decision making.关键词：open-source intelligence fusion;intelligence spoofing attack;intelligence anti-spoofing defense;deepfake;public opinion detection185|43|0更新时间：2024-09-15
- 摘要：As a new paradigm in the field of artificial intelligence, game reinforcement learning is an advanced mainstream method to solve the edge computing problem of low-orbit constellation. The multi-agent deep reinforcement learning integrated into the game perspective provides a new idea for dynamic, complex and uncertain constellation edge computing problems. By summarizing the three main research directions of satellite edge computing, namely satellite networking, task unloading and resource scheduling, the basis of game reinforcement learning paradigm is elaborated, and the typical applications in the three research directions are described respectively from the methods of game model, deep Q network, deep deterministic strategy gradient and near-end strategy optimization. In the end, the paper looks forward to the frontier challenges in this field, expected to provide a reference for the cross-fusion research of game reinforcement learning paradigm and low-orbit constellation edge computing.关键词：low-orbit constellation;edge computing;game theory;multi-agent reinforcement learning82|31|0更新时间：2024-09-15
Surveys and Prospectives
- 摘要：The traditional thoracolumbar fracture image-assisted classification method has low accuracy and poor generalization ability. Therefore, based on deep convolutional neural network, this paper proposes an AI classification method for thoracolumbar fracture for auxiliary diagnosis. Firstly, a total of 698 CT images of patients with lumbar spine fractures were collected from Sichuan Integrative Medicine Hospital, and a data set was established, including 279 compression fractures (category A), 295 burst fractures (category B), and 124 normal (category C). Secondly, the convolutional neural network model ResNet-50 was modified and combined with transfer learning to train, verify and test the data set to obtain the AI classification model of thoracolumbar fracture. Then, the Confusion Matrix is used to evaluate the prediction model. The accuracy of the training set and the validation set of the model is 95.75% and 96.36%, respectively, indicating that the model obtained by training has good accuracy and generalization ability. This paper proposes an image-assisted classification method for thoracolumbar fracture, which can improve the efficiency and accuracy of manual diagnosis.关键词：thoracolumbar fracture;deep convolutional neural network;AI classification method;generalization ability80|38|0更新时间：2024-09-15
- 摘要：It had been demonstrated that the functional brain networks of Alzheimer’s patients exhibit topological properties distinct from those of cognitively normal individuals, including small-world and clustering coefficients. To further elucidate the structural characteristics of the brain networks of Alzheimer’s patients and their differences in brain network structure with cognitively normal individuals, fractal features were introduced for comparative analysis. Experimental results demonstrated a significant difference in the minimum number of boxes and connection ratio between brain networks of Alzheimer’s patients and cognitively normal individuals at the same box-covering radius. At the brain network threshold of 0.4, the connection ratio values of cognitively normal individuals’ brain networks were distributed in the [0,1] interval, while the connection ratio values of Alzheimer’s patients’ brain networks were distributed in the (0.5,1) interval. Furthermore, the average connection ratio of Alzheimer’s patients’ brain networks was found to be higher than that of cognitively normal individuals at different thresholds. These differences in network structure provide a new reference for the diagnosis of Alzheimer’s disease.关键词：Alzheimer’s disease;fractal feature;box-covering method;connection ratio67|30|0更新时间：2024-09-15
- 摘要：This paper proposes OrthoGPT, a multimodal generative pre-trained transformer model designed for precise diagnosis and treatment of orthopedic diseases. It addresses several challenges in orthopedic care, such as the difficulties in emergency management outside hospital settings, the complexity and high risks associated with surgical planning, and the strong demand for personalized rehabilitation therapy. The model leverages multimodal feature alignment and fusion technology, multi-scenario personalized diagnosis and treatment plan generation technology, combined with an orthopedic world model based on parallel theory and multi-agent methods. It aims to provide preliminary diagnosis and emergency guidance in non-hospital scenarios, simulate and predict personalized postoperative treatment outcomes, assist doctors in surgical planning, and offer personalized rehabilitation treatment suggestions. The purpose of this paper is to explore the potential applications of OrthoGPT in improving the efficiency and effectiveness of orthopedic diagnosis and treatment, enhancing the patient experience, and advancing the development and application of artificial intelligence in the field of orthopedics.关键词：orthopedics;large language models;smart healthcare;parallel medicine202|52|0更新时间：2024-09-15
- 摘要：The superior data analytics and logical reasoning capabilities of big language models provide new ideas for real-time traffic management and assisted decision-making. ChatGPT efficiently processes and analyzes publicly available social media data to detect city and roadway information and traffic events contained in the data, which can be used to assist traffic managers in making real-time inquiries, tracing causes and exploring countermeasures. This paper constructs an intelligent Q&A framework, TMGPT (traffic management GPT), which integrates social media data with ChatGPT, to explore how large language models can be leveraged to quickly detect traffic anomalies and provide decision support for traffic management departments. Through the acquisition, processing and analysis of social media data, the framework achieves accurate detection of traffic anomalies and the generation of targeted response strategies, and continuously optimizes the system performance through the feedback mechanism, providing a decision basis for traffic management and policy-making departments to improve the efficiency and safety of urban traffic operation. The results show that compared with traditional methods, TMGPT significantly improves the accuracy of detection and reduced response time in the detection and assisted decision-making of abnormal traffic events, which demonstrates the application potential of large language models in complex urban traffic management.关键词：ChatGPT;traffic event detection;social media;auxiliary decision-making249|66|0更新时间：2024-09-15
- 摘要：Air quality forecasting is an effective means of managing and mitigating air pollution. To enhance prediction accuracy, a new air quality prediction model, namely the sample convolutional and interaction network(SCINet), is introduced in this paper. The model is composed of multiple SCI-Blocks arranged in a complete binary tree structure. Whereafter, the time series is rearranged through flipping odd-even splits, and a new sequence is generated,which is able to capture the complex dependencies and local trends of multivariate atmospheric pollutants better. Given the seasonality and randomness of monitoring data for atmospheric pollutant, the paper employs two SCINets for stacking, which not only expands the receptive field of convolutional operations, but also enables multi-resolution analysis. Furthermore, through the optimization of model depth and hyperparameters, the model may fit the temporal characteristics of air quality time series data better, which is helpful to extract the temporal relationship features of the target variable. In the end, the Beijing PM2.5 dataset and the Beijing multi-site air quality dataset are utilized to evaluate the effectiveness of SCINet. The results show that SCINet has higher prediction accuracy, whose the root mean square error (δRMSE) is reduced by 31.59% in short-term prediction and 24.36% in long-term prediction compared with the best-performing DAQFF model.关键词：air quality forecasting;SCINet;CNN;time series75|36|1更新时间：2024-09-15
- 摘要：The detection of train track defects is of great significance to the safety of rail transportation, but current manual inspection can no longer meet the complex and heavy track inspection requirements. Deep learning methods have greatly expanded the means and detection capabilities of defect detection, and in order to improve the efficiency and quality of surface defect detection, the current trends of visual detection methods in conjunction with the types and related attributes of track defects was systematically analyzed. The basic principles, technologies, methods, and current application status of visual detection methods for track and fastening defects, as well as the concepts, applications, and significance of these detection methods were elaborated. Finally, the current trends in the field of track defect detection are analyzed and summarized, and for the first time proposes the concept of a fully automatic track maintenance system, which aims to provide useful support and reference for future related research.关键词：rail transportation safety;rail detect;image recognition;deep learning125|44|0更新时间：2024-09-15
Papers and Reports
- 摘要：In response to the potential risk of backdoor attacks faced by artificial intelligence systems, a range of backdoor defense strategies are developed. The diversity of the evaluation criteria for existing defense method, makes cross-method comparisons a significant challenge. Hence, a unified evaluation framework base on artificial intelligence backdoors was proposed. This framework aimed to provide a common standard for evaluating different levels of defense strategies, including dataset-level and model-level defenses. Regarding the dataset-level defense strategies, the effectiveness of backdoor detection was primarily assessed through accuracy. Regarding the model-level defense strategies, focus was mainly placed on metrics such as attack success rate. By implementing unified evaluation framework, the performance of various backdoor defense methods under the same standards were compared and analyzed. This not only aids in identifying the strengths and weaknesses of each method, but also proposes targeted suggestions for improvements. The results indicate that unified evaluation framework can effectively measure the performance of different defense strategies, providing an important reference for further enhancing the security of artificial intelligence systems.关键词：artificial intelligence security;backdoor attack;backdoor defense;unified evaluation67|38|0更新时间：2024-09-15
- 摘要：Large language models(LLM) are leading to a new revolution in artificial intelligence technology. As LLM have the characteristics of platform, mass data dependence, frequent user interaction, and easy to be attacked, the risk of privacy disclosure is more worrying. The source and internal mechanism of privacy leakage caused by LLM were focused. After reviewing the domestic and foreign experience in the governance of LLM privacy risks, a five-dimensional governance framework that included policies, standards, data, technology, and ecology were proposed. Finally, this paper also looks forward to the new privacy disclosure risks that LLM may face under the development trend of multimodal, agent, embodied intelligence, edge intelligence, etc.关键词：large language model;privacy;data leakage;government75|33|0更新时间：2024-09-15
- 摘要：Agile governance, as a new paradigm of governance, is conducive to enhancing the governance level of generative artificial intelligence. By analyzing the governance needs and development challenges of generative artificial intelligence, the inadequacies of traditional governance models were revealed. It was believed that agile governance, characterized by multi-stakeholder co-governance, flexible governance basis, and iterative processes, could effectively balance the needs of technological innovation and risk control, and compensate for the limitations of traditional governance paradigms in the governance of generative artificial intelligence. After examining the international practices of agile governance, suggestions for agile governance strategies suitable for generative artificial intelligence in China were proposed. To ensure the high-quality development and high-level safety of generative artificial intelligence in China, it is necessary to further develop a multi-stakeholder co-governance model on the basis of multi-participation, strengthen the interaction between hard law and soft law, and gradually enrich the governance toolkit of generative artificial intelligence.关键词：generative artificial intelligence;agile governance;co-governance;deregulation;iterative process109|46|1更新时间：2024-09-15
Special Column: Artificial Intelligence Trustworthiness and Governance
- Postal code：100079
- Tel：（010）53879206 Email：cjist@bjxintong.com.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
- Total Visits：161982 Visits Today：145
0