Search

Abstract Aviation emergency management is playing a more and more important role in the aviation field. How to make effective use of massive heterogeneous and multi-source aviation accident knowledge has become a great challenge for aviation emergency management. Aiming at the problems such as too long physical length, mixed and composite entities, similar character of domain entity names, information difference between entities, separation of codes between entities, coding errors during transmission, etc., the construction method of knowledge map of civil aviation emergencies is studied. In previous research methods, entity link is always divided into two parts, that is, first detection and then disambiguation, which makes the mentioned entity and the candidate entity are encoded separately, and there is error transmission between the two parts, modules cannot communicate with each other, and the close association between entities cannot be well learned. In this paper, we proposed an end-to-end entity linking method based on two-layer BiLSTM model joint coding vectorize each word of civil aviation text information, and then concatenate feature vectors into two-layer BiLSTM model to obtain high-level context representation. Because the joint encoding of boundary information can reduce the error transmission, information is exchanged between candidate entities during the initial encoding to enhance the closeness between candidate entities and candidate entities. The experimental results show that compared with other sota models, the F1 value of the proposed model reaches 88.97%.

Abstract Due to their high flexibility, low cost, and ease of handling, Unmanned Aerial Vehicles (UAVs) are often used to perform difficult tasks in complex environments. Stable and reliable path planning capability is the fundamental demand for UAVs to accomplish their flight tasks. Most researches on UAV path planning are carried out under the premise of known environmental information, and it is difficult to safely reach the target position in the face of unknown environment. Thus, an autonomous collision-free path planning algorithm for UAVs in unknown complex environments (APPA-3D) is proposed. An anti-collision control strategy is designed using the UAV collision safety envelope, which relies on the UAV's environmental awareness capability to continuously interact with external environmental information. A dynamic reward function of reinforcement learning combined with the actual flight environment is designed and an optimized reinforcement learning action exploration strategy based on the action selection probability is proposed. Then, an improved RL algorithm is used to simulate the UAV flight process in unknown environment, and the algorithm is trained by interacting with the environment, which finally realizes autonomous collision-free path planning for UAVs. The comparative experimental results in the same environment show that APPA-3D can effectively guide the UAV to plan a safe and collision-free path from the starting point to the target point in an unknown complex 3D environment.

In reinforcement learning, off-policy temporal difference learning methods have gained significant attention due to their flexibility in utilizing existing data. However, traditional off-policy temporal difference methods often suffer from poor convergence and stability when handling complex problems. To address these issues, this paper proposes an off-policy temporal difference algorithm with Bellman residuals (TDBR). By incorporating Bellman residuals, the proposed algorithm effectively improves the convergence and stability of the off-policy learning process. This paper first introduces the basic concepts of reinforcement learning and value function approximation, highlighting the importance of Bellman residuals in off-policy learning. Then, the theoretical foundation and implementation details of the TDBR algorithm are described in detail. Experimental results in multiple benchmark environments demonstrate that the TDBR algorithm significantly outperforms traditional methods in terms of both convergence speed and solution quality. Overall, the TDBR algorithm provides an effective and stable solution for off-policy reinforcement learning with broad application prospects. Future research can further optimize the algorithm parameters and extend its application to continuous state and action spaces to enhance its applicability and performance in real-world problems.

This paper presents a novel multi-dimensional asymmetric game model for network attack–defense decision-making, called “Catch the Cyber Thief”. The model is built upon the concept of partially observable stochastic games (POSG) and is designed to systematically incorporate multi-dimensional asymmetry into network attack–defense problems. The attack agent is called a “thief” who wants to control a key host by exploring the unfamiliar network environment, and the defense agent is called a “police” who needs to catch the opponent before its goal is accomplished. The results indicate that the asymmetry of network attack and defense is not only influenced by attack and defense strategies but also by spatio-temporal factors such as the attacker’s initial position, network topology, and defense lag time. In addition, we have found that there may exist the “attack rhythm,” which makes “how to maintain a good attack rhythm” and “how to generate a robust defense strategy against different attackers” worth exploring. Compared with existing attack–defense game models, our game model can better generate a direct mapping relationship with real elements, enabling us to understand network attack and defense interactions better, recognize security risks, and design defense strategies that can directly serve real-world decision-making.

In the process of building domain knowledge graph, the result of relationship extraction between entities is an important guarantee of the quality of the graph. Therefore, we propose a clustering method based on reinforcement learning for remote supervised relation extraction. For the relationship extraction of accident information in the aviation domain mapping, a clustering method combining local dense and global dissimilarity is proposed in combination with remote supervision, which can obtain a large amount of low-noise labeled data and reduce part of the wrong labeling and missing labeling due to the strong specialization in the aviation domain; meanwhile, reinforcement learning is introduced to denoise the negative instance noise in the positive sample data; Finally, we propose a two-attention segmentation (DAPCNN) relationship extraction model to mine deep semantic sentences. The experimental results show that in the civil aviation relationship extraction text constructed in this paper, the Micro_R, Micro_P and Micro_F1 values of the proposed relationship extraction method reach 83.41 %, 84.16 % and 83.96 %. In the open relationship extraction dataset DuIE, The Micro_R, Micro_P and Micro_F1 of the proposed method are up to 83.41 %, 93.58 % and 94.02 % respectively. Compared with the current advanced multi-instance and multi-label model, the proposed method can more accurately extract the relationship between aviation accident entities. At the same time, the performance of the open data set is also good, and has a certain universality.

Off-policy is a key setting for reinforcement learning algorithms. In recent years, the stability of off-policy learning for value-based reinforcement learning has been guaranteed even when combined with linear function approximation and bootstrapping. Convergence rate analysis is currently a hot topic. However, the convergence rates of learning algorithms vary, and analyzing the reasons behind this remains an open problem. In this paper, we propose an essentially simplified version of a convergence rate to generate general off-policy temporal difference learning algorithms. We emphasize that the primary determinant influencing convergence rate is the minimum eigenvalue of the key matrix. Furthermore, we conduct a comparative analysis of the influencing factor across various off-policy learning algorithms in diverse numerical scenarios. The experimental findings validate the proposed determinant, which serves as a benchmark for the design of more efficient learning algorithms.

Due to the limited understanding of the physical and chemical processes involved in ozone formation, as well as the large uncertainties surrounding its precursors, commonly used methods often result in biased predictions. Deep learning, as a powerful tool for fitting data, offers an alternative approach. However, most deep learning-based ozone-prediction models only take into account temporality and have limited capacity. Existing spatiotemporal deep learning models generally suffer from model complexity and inadequate spatiality learning. Thus, we propose a novel spatiotemporal model, namely the Spatiotemporal Attentive Gated Recurrent Unit (STAGRU). STAGRU uses a double attention mechanism, which includes temporal and spatial attention layers. It takes historical sequences from a target monitoring station and its neighboring stations as input to capture temporal and spatial information, respectively. This approach enables the achievement of more accurate results. The novel model was evaluated by comparing it to ozone observations in five major cities, Nanjing, Chengdu, Beijing, Guangzhou and Wuhan. All of these cities experience severe ozone pollution. The comparison involved Seq2Seq models, Seq2Seq+Attention models and our models. The experimental results show that our algorithm performs 14% better than Seq2Seq models and 4% better than Seq2Seq+Attention models. We also discuss the interpretability of our method, which reveals that temporality involves short-term dependency and long-term periodicity, while spatiality is mainly reflected in the transportation of ozone with the wind. This study emphasizes the significant impact of transportation on the implementation of ozone-pollution-control measures by the Chinese government.

Background: Type 2 diabetes mellitus (T2DM) is a multifaceted disorder affecting epidemic proportion at global scope. Defective insulin secretion by pancreatic β-cells and the inability of insulin-sensitive tissues to respond effectively to insulin are the underlying biology of T2DM. However, circulating biomarkers indicative of early diabetic onset at the asymptomatic stage have not been well described. We hypothesized that global and targeted mass spectrometry (MS) based metabolomic discovery can identify novel serological metabolic biomarkers specifically associated with T2DM. We further hypothesized that these markers can have a unique pattern associated with latent or early asymptomatic stage, promising an effective liquid biopsy approach for population T2DM risk stratification and screening.Methods: Four independent cohorts were assembled for the study. The T2DM cohort included sera from 25 patients with T2DM and 25 healthy individuals for the biomarker discovery and sera from 15 patients with T2DM and 15 healthy controls for the testing. The Pre-T2DM cohort included sera from 76 with prediabetes and 62 healthy controls for the model training and sera from 35 patients with prediabetes and 27 healthy controls for the model testing. Both global and targeted (amino acid, acylcarnitine, and fatty acid) approaches were used to deep phenotype the serological metabolome by high performance liquid chromatography-high resolution mass spectrometry. Different machine learning approaches (Random Forest, XGBoost, and ElasticNet) were applied to model the unique T2DM/Pre-T2DM metabolic patterns and contrasted with their effectiness to differentiate T2DM/Pre-T2DM from controls.Results: The univariate analysis identified unique panel of metabolites (n = 22) significantly associated with T2DM. Global metabolomics and subsequent structure determination led to the identification of 8 T2DM biomarkers while targeted LCMS profiling discovered 14 T2DM biomarkers. Our panel can effectively differentiate T2DM (ROC AUC = 1.00) or Pre-T2DM (ROC AUC = 0.84) from the controls in the respective testing cohort.Conclusion: Our serological metabolite panel can be utilized to identifiy asymptomatic population at risk of T2DM, which may provide utility in identifying population at risk at an early stage of diabetic development to allow for clinical intervention. This early detection would guide ehanced levels of care and accelerate development of clinical strategies to prevent T2DM.

In recent years, with the explosive growth of the data, a wide range of data in Cyber-Physical-Social Systems (CPSS) are generated and collected as big data. Cloud computing have been widely-used as the supporting computation infrastructure, which makes big data analysis gaining much attention from IT industry and academia. Moreover, the data often are distributed and stored in different computation resources in many big data applications. Therefore, distributed computing and optimization has been developed for solving big data problems in cloud computing. time efficiency is the significant bottleneck in the performance of distributed optimization algorithms. In this paper, we propose a novel fast distributed algorithm via Alternating Direction Method of Multipliers with Adaptive Local Update (ADMM-ALU), that uses an efficient adaptive local update strategy to accelerate the speed of convergence by automatically determining the number of inner iterations of local update in each outer iteration (communication round). In particular, our method applies the optimality conditions and the magnitudes of residuals of ADMM to freely steer the trade-off between communication and local computation. Empirically, the performance of our method is tested on several benchmark datasets, and the experimental results show that compared to various versions of ADMM algorithms, our method converges faster, and could be a highly effective and efficient algorithm for practical use in big data applications.

In recent years, cyber attacks have shown diversified, purposeful, and organized characteristics, which pose significant challenges to cyber defense decision-making on internal networks. Due to the continuous confrontation between attackers and defenders, only using data-based statistical or supervised learning methods cannot cope with increasingly severe security threats. It is urgent to rethink network defense from the perspective of decision-making, and prepare for every possible situation. Reinforcement learning has made great breakthroughs in addressing complicated decision-making problems. We propose a framework that defines four modules based on the life cycle of threats: pentest, design, response, recovery. Our aims are to clarify the problem boundary of network defense decision-making problems, to study the problem characteristics in different contexts, to compare the strengths and weaknesses of existing research, and to identify promising challenges for future work. Our work provides a systematic view for understanding and solving decision-making problems in the application of reinforcement learning to cyber defense.

Landslides are one of the most frequent and serious geological disasters that threaten people’s lives and property safety. In recent years, with the rapid development of the coastal economy and the increasingly strained spatial resources, the island development activities have become extremely rapid, resulting in the frequent occurrence of landslides on the island. We selected Beichangshan Island in the north of China as the research area. By using high-precision ground-based real aperture radar (GB-RAR) measurement technology, the displacement changes of potential landslides are monitored continuously and dynamically to realize the real-time diagnosis and early warning of island landslides. At the same time, the data interpretation method and key processing flow are described in detail. The results show that during the whole monitoring process, an area of obvious change is found, which is mainly located in the middle of the landslide mass. The mean velocity rate shows a nonlinear deformation trend. The maximum deformation of the landslide in the five selected points reaches 4.5 mm, which indicates that the area is in an unstable stage. The deformation monitoring ability of GB-RAR technology to identify the sub-millimeter level is demonstrated, and the monitoring method is verified. The validity and reliability of the method can be applied to real-time dynamic fine deformation diagnosis of island landslides. Its accuracy can meet the needs of dynamic change monitoring of island landslides, and it can become an important tool and means for early warning and treatment of landslides. The research is conducive to further enriching and improving the monitoring method system of island geological disasters in China, provides a scientific basis and technical support for early warning and disaster prevention and mitigation of island landslides, and can be popularized and applied in the monitoring of island landslides.

Objective To investigate the association of the CD11b single nucleotide polymorphism (SNP) rs1143679 with systemic lupus erythematosus (SLE) in Han Chinese patients, and to clarify this association with SLE clinical manifestations. Methods PCR–restriction fragment length polymorphism and direct sequencing of CD11b rs1143679 were conducted in 584 patients with SLE and 628 healthy controls in this case–control study to compare genotype and allele frequency distributions. Correlations between CD11b genotypes and clinical manifestations were also determined. Results The frequency of the CD11b rs1143679 GA genotype was 1.89% in Han Chinese patients with SLE, which was much lower than that of European and American populations, but close to the frequency observed in individuals from Hong Kong and Thailand. The CD11b rs1143679 GA genotype was also shown to confer susceptibility to SLE (odds ratio = 4.00, 95% confidence interval = 1.11–14.41). CD11b rs1143679 was found to be significantly associated with nephritis, but not with age of disease onset, arthritis, hematological involvement, or neural lesions. Conclusion CD11b rs1143679 appears to be associated with risk for SLE in the Han Chinese population, and may play an important role in the development of lupus nephritis.

The processing capacity and power of nodes in a Wireless Sensor Network (WSN) are limited. And most image compression algorithms in WSN are subject to random image content changes or have low image qualities after the images are decoded. Therefore, an image compression method based on multilayer Restricted Boltzmann Machine (RBM) network is proposed in this paper. The alternative iteration algorithm is also applied in RBM to optimize the training process. The proposed image compression method is compared with a region of interest (ROI) compression method in simulations. Under the same compression ratio, the qualities of reconstructed images are better than that of ROI. When the number of hidden units in top RBM layer is 8, the peak signal-to-noise ratio (PSNR) of the multilayer RBM network compression method is 74.2141, and it is much higher than that of ROI which is 60.2093. The multilayer RBM based image compression method has better compression performance and can effectively reduce the energy consumption during image transmission in WSN.

In this study, we mapped the co-seismic deformation of the 2015 Mw 8.3 Illapel, Chile earthquake with multiple Sentinel-1A TOPS data frames from both ascending and descending geometries. To meet the requirement of very high co-registration precision, an improved spectral diversity method was proposed to correct the co-registration slope error in the azimuth direction induced by multiple Sentinel-1A TOPS data frames. All phase jumps that appear in the conventional processing method have been corrected after applying the proposed method. The 2D deformation fields in the east-west and vertical directions are also resolved by combing D-InSAR and Offset Tracking measurements. The results reveal that the east-west component dominated the 2D displacement, where up to 2 m displacement towards the west was measured in the coastal area. Vertical deformations ranging between −0.25 and 0.25 m were found. The 2D displacements imply the collision of the Nazca plate squeezed the coast, which shows good accordance with the geological background of the region.