Loading...

Table of Content

      
    20 January 2024
    Volume 59 Issue 1
    Invited Review
    Strategic limit theory and strategic statistical learning
    Xiaodong YAN
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  1-10, 45.  doi:10.6040/j.issn.1671-9352.0.2023.512
    Abstract ( 354 )   HTML ( 9 )   PDF (2306KB) ( 454 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The nonlinear expectation is an original research direction pioneered by Academician Peng Shige of Shandong University, which is becoming increasingly important in various fields of scientific research. The rise of big data and artificial intelligence has provided stronger impetus for innovative theoretical and applied research in nonlinear expectation. Recently, Shandong University's Nonlinear Probability Team has developed the "Strategy Limit Theory" based on the strategic game process of multi-armed bandits, representing a significant breakthrough in the intersection of nonlinear probability theory and reinforcement learning. This has tran-sformed the research paradigm of traditional statistical methods. Based on the proposed 10 basic mathematical problems of artificial intelligence by Academician Xu Zongben, the declaration guide of 2022 major research plan projects issued by the National Natural Science Foundation of China for the research about universal and interpretable artificial intelligence technologies, and the application guide for basic mathematical theory research of artificial intelligence in 2021 and 2022 the key projects of "Mathematics and Applied Research" issued by the Ministry of Science and Technology, this article adopts the concept of "strategy" to reveal the nature of artificial intelligence and explore and the motivation source and theoretical basis for initiating and promoting the innovation of artificial intelligence technology. Different from the applications of the traditional law of large numbers and the central limit theorem in the field of artificial intelligence, we propose novel theory about the strategic law of large numbers and the central limit theorem in the new generation of artificial intelligence. The discussed topics in this work include but not limited to: (1) strategic sampling of massive data; (2) online learning of streaming data; (3) the central limit theorem of reinforcement learning; (4) differential privacy protection of data; (5) strategic integration of federal learning; (6) information reconstruction of transfer learning and meta learning; (7) the fusion of knowledge reasoning and data driving.

    Optimal monetary policy with a zero lower bound on the nominal interest rate under a continuous-time framework
    Haodong LIU,Chi ZHANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  11-16.  doi:10.6040/j.issn.1671-9352.0.2023.148
    Abstract ( 253 )   HTML ( 3 )   PDF (925KB) ( 122 )   Save
    References | Related Articles | Metrics

    In this paper, we give the continuous time version of the New Keynesian model, which is a backward stochastic differential system and translate the optimal monetary policy problem into a stochastic optimal control problem under control constraints. By using the maximum principle for the control system, we obtain the necessary condition for the optimal monetary policy. Also we give the expression of the optimal monetary policy.

    Prediction of average queue time in multi-server tandem queueing systems
    Yiran LI,Ning ZHAO,Zhijian ZHANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  17-26.  doi:10.6040/j.issn.1671-9352.4.2022.8254
    Abstract ( 312 )   HTML ( 3 )   PDF (1205KB) ( 480 )   Save
    Figures and Tables | References | Related Articles | Metrics

    This paper studies a multi-server tandem queueing system with two stations and infinite buffers before each station. The average queueing time of the two stations is predicted by linear regression models and nonlinear methods of machine learning, and the error in the prediction results of various machine learning methods is analyzed. Numerical experiments show that the nonlinear method exhibits better performance than the linear regression model. Moreover, the RF, XGBoost and GBDT methods are effective to predict the average waiting time of multi-server tandem queueing networks.

    Imagedata control chart based on nonnegative CP tensor decomposition
    Jin-yu FAN,Yang ZOU,Jian XIONG,Yongyi GU
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  27-34.  doi:10.6040/j.issn.1671-9352.0.2023.035
    Abstract ( 291 )   HTML ( 3 )   PDF (3049KB) ( 275 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Nonnegative tensor decomposition is well known for extracting the features of image data effectively and do not destroy the internal structure features of the data at the same time. This paper establishes a control chart for image data based on nonnegative tensor decomposition without any additional parameters. The monitoring performance of the proposed chart is verified by simulation under location changes, area changes, shape changes and color changes of images. Meanwhile, through a real industrial nonwoven fabric image, comparisons of the proposed control chart, the GLR-based spatiotemporal chart, the EWMA and region growing based chart and the RTC chart are conducted with the same parameter settings. The results show that our proposed method is superior to the other methods when shift size is less than or equal to 2 and performs similarly with the other method when the shift size is large than 2.

    Multi-instance embedding learning with instance affinity mining and reinforcement
    Mei YANG,Wen DENG,Benwen ZHANG,Fan MIN
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  35-45.  doi:10.6040/j.issn.1671-9352.4.2022.606
    Abstract ( 261 )   HTML ( 1 )   PDF (5354KB) ( 319 )   Save
    Figures and Tables | References | Related Articles | Metrics

    We propose the multi-instance embedding learning with instance affinity mining and reinforcement(MEMR) algorithm, including three techniques. The affinity mining technique is based on a custom affinity metric. First, the initial negative representative instance set(INRI) is selected in the negative instance space. Then, the initial positive representative instance set(IPRI) is chosen according to the difference between positive and negative instances. The affinity reinforcement technique evaluates the positive(negative) affinity between IPRI(INRI) and the entire instance space to obtain a representative instance set with stronger overall affinity. The bag embedding technique converts bags into single vectors for learning through the designed embedding function. Experiments are carried out across four application domains and seven comparison algorithms. The results show that MEMR generally outperforms other comparison algorithms in accuracy, especially in image retrieval and web recommendation datasets.

    Improved density peak clustering approach based on African vultures optimization algorithm
    Xinglong LUO,Xingshi HE,Jie ZHOU,Xinshe YANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  46-55,71.  doi:10.6040/j.issn.1671-9352.0.2022.623
    Abstract ( 280 )   HTML ( 2 )   PDF (4733KB) ( 126 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Density peak clustering algorithm is a new fast search algorithm for automatically finding cluster centers. Aiming at the uncertainty of its cut-off distance and the instability of the one-step allocation strategy, an improved density peak clustering approach based on African vultures optimization algorithm is proposed. The objective function of the optimization problem is established through evaluating accuracy(Acc), and the uncertain cut-off distance dc is optimized by the powerful optimization ability of the African vultures optimization algorithm, which reduces the inaccuracy of artificial values. Secondly, according to the average density of the data set, it is divided into different density areas, and different allocation strategies are used for different areas. For data points in the high-density area, the same allocation method as the original density peak clustering is used, and for data points in the low-density area, the k-nearest neighbor method is used for clustering. Finally, the algorithm is experimentally verified on synthetic and real data sets, the clustering performance of the algorithm has been greatly improved, and the division of data sets with large density differences is also more accurate.

    A new minimal determinization method of nondeterministic fuzzy finite automata
    Ping LI,Jufang YANG,Yanping YANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  56-61.  doi:10.6040/j.issn.1671-9352.0.2022.596
    Abstract ( 276 )   HTML ( 1 )   PDF (442KB) ( 324 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The minimal determinization of nondeterministic fuzzy finite automata (NFFA) is an important problem in automata theory. A new minimal determinization method for nondeterministic fuzzy finite automata with membership values in lattice-ordered monoids called interior construction is presented. For this reason, the definition of the interior of a fuzzy state and its related properties are given firstly. It is further proved that for any nondeterministic fuzzy finite automata, a minimal deterministic fuzzy finite automata is obtained by using the internal properties of fuzzy states, which is equivalent to it, finally, the correctness of this method is verified by an example.

    Power allocation algorithm for CR-NOMA based on adaptive bacterial foraging optimization strategy
    Yi PENG,Xiaolin MA,Qingqing YANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  62-71.  doi:10.6040/j.issn.1671-9352.0.2023.043
    Abstract ( 256 )   HTML ( 4 )   PDF (2021KB) ( 193 )   Save
    Figures and Tables | References | Related Articles | Metrics

    A power allocation algorithm based on an adaptive bacterial foraging optimization strategy is proposed to aim at the problem of low spectrum utilization of cognitive radio non-orthogonal multiple access (CR-NOMA) system with underlay mode in multiple primary and secondary user scenarios, Firstly, the joint user matching is carried out, and the secondary user grouping problem is equivalent to the secondary user-subchannel bidirectional dynamic matching problem. Secondly, the power scale factor vector of the secondary user is constructed and mapped into the position vector of the bacterial individual, and the bacterial swimming step and rotation direction are improved in the trend operation. In the replication operation, the differential evolution algorithm is used to perform mutation selection on the first half of the high-quality solutions. In the migration operation, the migration range is defined, and the adaptive migration probability is used to speed up the process of finding the best position vector. Finally, the optimal power scaling factor is obtained to maximize the total throughput of the system. The results show that compared with the hierarchical pairing power allocation (HPPA) algorithm and the CR-OMA algorithm, the proposed algorithm can effectively accelerate the convergence speed, enhance the global optimization ability, and have better system performance.

    Dynamic evolution of E-supply chain efficiency and fairness degree in the mixed-sale mode underfairness concern
    Min XIAO,Xiaodian ZHANG,Xinhua HE
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  72-84.  doi:10.6040/j.issn.1671-9352.0.2022.662
    Abstract ( 290 )   HTML ( 0 )   PDF (1037KB) ( 85 )   Save
    Figures and Tables | References | Related Articles | Metrics

    An E-supply chain composed of a manufacturer and an e-commerce platform with self-established logistics was established. The manufacturer not only sells their products through the e-commerce platform agent, but also wholesales products to the e-commerce platform for distribution. From the perspective of the e-commerce platform′ fairness concern about the dynamic change of information, this paper analyzes the impact of fairness concern about the dynamic change of information on the pricing decisions, efficiency and fairness degree of the E-supply chain under the mixed-sale mode. The research shows that when the e-commerce platform transmits the true fairness concern information or false fairness concern information to the manufacturer, it is beneficial to obtain more profit for itself, but the intensity value of fairness concern should be within a reasonable range, otherwise, it will reduce the subjective fairness degree and objective fairness degree of the supply chain. When the e-commerce platform does not transmit the true fairness concern information to the manufacturer, the profit of each member of the supply chain and the total profit of the supply chain are the lowest, and the efficiency of the supply chain will decrease with the increase of fairness concern. When the proportion of revenue sharing obtained by the manufacturer is within a certain range, it is beneficial to improve the subjective fairness degree and objective fairness degree of the supply chain. However, when it exceeds a certain range, it will reduce the fairness degree of the supply chain.

    Government subsidy strategy of low carbon supply chain considering retailer's corporate social responsibility
    Feng LI,Chun-long CHENG,Ye-feng GUO
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  85-99.  doi:10.6040/j.issn.1671-9352.0.2022.412
    Abstract ( 324 )   HTML ( 1 )   PDF (1835KB) ( 343 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Considering that retailers have corporate social responsibility(CSR), a tripartite game model consisting of government, manufacturers, and retailers was established to compare and analyze the social welfare, carbon emission reduction rate, demand for low-carbon products, and environmental improvement under four different subsidy strategies: no government subsidy, R&D subsidy, production subsidy, and a dual subsidy which composed of R&D subsidy and production subsidy, the optimal subsidy strategy of the government was studied. The results indicate that the improvement of consumer low-carbon preferences and research and development efficiency will promote environmental improvement and increase product emission reduction rates, demand for low-carbon products, corporate profits, and social welfare levels. As the CSR level of retailers increases, the government will reduce subsidies, but the overall social welfare, product emission reduction rate and demand for low-carbon products will not decrease. All four subsidy strategies are conducive to increasing carbon emission reduction rates, demand for low-carbon products, social welfare levels, and promoting environmental improvement, thereby better achieving economic, social, and environmental goals. Manufacturers prefer the government's dual subsidy strategy, while retailers prefer the R&D subsidy policy when their CSR level is high, otherwise, they prefer the dual subsidy strategy.

    Platform supply chain network operation decision based on blockchain technology under cap-and-trade regulation
    Cong SHI,Guitao ZHANG,Xiao ZHANG,Shuaicheng LIN
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  100-114, 123.  doi:10.6040/j.issn.1671-9352.0.2023.049
    Abstract ( 311 )   HTML ( 7 )   PDF (4357KB) ( 388 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Under the cap-and-trade regulation, based on Nash non-cooperative game theory and variational inequality theory, considering supplier's misrepresentation of carbon caps during the initial allocation and the demand of green information sensitive consumers for product verification, two platform supply chain network equilibrium models with and without blockchain technology are constructed. Then this paper explores and compares the equilibrium conditions of supply chain members and blockchain technology application thresholds before and after the application of blockchain technology. The modified contraction projection algorithm is used to solve the problem and some numerical examples are given to show that: the application of blockchain technology can avoid the inconsistency of carbon emission trading volume, further reduce the waste of resources and improve the profit of the supply chain. The supplier's misrepresentation of the carbon cap can to some extent improve the overall profit of the supply chain, but it inhibits the ability of the supply chain to achieve higher profits. Within a certain threshold of blockchain technology operating costs, supply chain members have the motivation to adopt blockchain technology.

    Interval time series models-based vessel price index forecasting
    Yuanyuan XU,Hongyue GUO,Lidong WANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  115-123.  doi:10.6040/j.issn.1671-9352.4.2022.205
    Abstract ( 305 )   HTML ( 1 )   PDF (2108KB) ( 338 )   Save
    Figures and Tables | References | Related Articles | Metrics

    This study considers the forward freight agreement (FFA) as an exogenous variable to analyze its specific impact on the vessel price index. With the center and range method, an interval autoregression model, an interval error correction model considering the co-integration between the upper and lower bounds of the interval time series, and interval error correction model with an additional incorporation of the exogenous variable FFA are established, respectively. The constructed models are employed to perform the interval prediction of the bulk carrier newship price index and bulk carrier secondhand price index. Based on the criteria MAE and RMSE, the prediction accuracy is higher after adding the co-integration term and FFA into the models.

    Delay optimal control of 1,3-propanediol batch fermentation
    Xiao WANG,Chongyang LIU,Dianzhong HU,Gang LIU
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  124-131, 138.  doi:10.6040/j.issn.1671-9352.0.2022.542
    Abstract ( 288 )   HTML ( 5 )   PDF (788KB) ( 303 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In the batch process of glycerol bioconversion to 1,3-propanediol, the initial biomass and glycerol concentration will affect the productivity of 1,3-propanediol. This paper proposes a constrained delay optimal control model to maximize the productivity of 1,3-propanediol. For this problem, the time-scaling transformation is applied to convert it to an optimal control problem with fixed terminal time. Then, the penalty method is used to deal with the constraints in the optimal control problem. Finally, a hybrid algorithm of simulated annealing and genetic algorithm is developed to solve the resulting problem. Numerical results indicate that the productivity of 1,3-propanediol increases by 21.12% compared with the previous result.

    Differences in height, diameter at breast height, and growth relationships between them of common tree species at different altitudes in Shandong Province
    Wenxin ZHANG,Qiang LI,Ning WANG,Xiaoli FAN,Hui WANG,Chengping JIANG,Yu LIANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  132-138.  doi:10.6040/j.issn.1671-9352.0.2023.216
    Abstract ( 404 )   HTML ( 5 )   PDF (1032KB) ( 252 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Based on the data of plant community inventory in Shandong Province, the common tree species were selected as the research objects to analyze the characteristics of diameter at breast height(DBH) and tree height. Single factor analysis of variance was used to compare whether DBH and tree height were significantly different at different altitudes. The allometric growth relationship between tree height and DBH of different tree species at different altitudes was studied by standardized major axis estimation. The study found that there were significant differences in DBH, tree height and their growth relationship at different altitudes, and height-DBH allometry of different tree species at different altitudes was different. The height-DBH growth relationship of three broad-leaved tree species, Quercus acutissima, Robinia pseudoacacia and Quercus variabilis, shows allometry growth at different altitudes, and the growth rate of DBH is greater than that of tree height. At low altitude, the growth rate of DBH of Platycladus orientalis was higher than that of tree height, while at middle altitude, the growth rate of tree height was higher than that of DBH. For Pinus tabulaeformis, the growth rate of tree hight is faster at low altitude areas, while the growth rate of DBH is faster in middle and high altitude areas. Pinus densiflora and Larix kaempferi have adopted isokinetic growth patterns in high and medium altitude areas. The environmental changes caused by altitude gradients can affect the growth process of trees, but the differences in growth response of different tree species are the result of the combined effects of habitat and genetic characteristics.

    Growth and physiological mechanism study of H. crustuliniforme and Lonicera japonica symbiotic relationship under salt tolerance
    Deyu MU,Nan LI,Tao WANG,Jin LIU,Zongzhao MU,Jingchuan ZHANG
    JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2024, 59(1):  139-150.  doi:10.6040/j.issn.1671-9352.0.2023.116
    Abstract ( 257 )   HTML ( 1 )   PDF (2682KB) ( 358 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Lonicera japonica and Hebeloma crustuliniforme were used as materials in this study. 0, 60, 90 and 120 mmol/L NaCl salt treatments were taken out to study the rule of H. crustuliniforme colonization in plant roots and analyze the mechanism of H. crustuliniforme on the growth and physiological characteristics of L. japonica seedlings. The results showed that with the increase in salt concentration, the proline content in the inoculated group plants was 35.98%, 141.28%, 71.66% and 8.79% higher than that of the control group, respectively. Under the treatment of 90 and 120 mml/L salt concentration, the salt tolerance of L.japonica plants was improved by the H. crustuliniforme. The results of correlation and two-way ANOVA indicated thatnew shoot growth and the number of branch growth were positively correlated with the biomass, and proline showed a highly significant negative correlation with the number of branch growth. H. crustuliniforme could significantly reduce the host seedling height growth and increase the host proline content significantly. The inoculation with H. crustuliniforme treatment could increase the proline accumulation of L. japonica seedlings to resist salt adversity under salt stress.