Search

Artificial intelligence (AI) images, like those produced by DALL-E, have seen explosive growth in the past several years and have the potential to disrupt numerous markets. While the technology offers exciting opportunities for creativity and innovation, it also raises important ethical considerations that must be addressed. These ethical implications include issues related to bias and discrimination, privacy, job displacement, and unintended consequences. To mitigate these challenges, a multi-disciplinary approach is needed, including the development of effective regulations and governance frameworks, the creation of unbiased algorithms, responsible data management practices, and educational and training programs. Additionally, encouraging ethical discussions and debates is crucial in ensuring the responsible use of AI-generated images. While AI-generated images offer many benefits, it is important to consider the ethical implications and work towards responsible AI practices to ensure their benefits are realized by society as a whole.

This paper proposes an improved cuckoo search (CS) algorithm combining nonlinear inertial weight and differential evolution algorithm (WCSDE) to overcome the shortcomings of the CS algorithm, such as low convergence accuracy, lack of information exchange within the population, and inadequate local search capabilities. Compared with other CS variants, two strategies are proposed in this paper to improve the properties of the WCSDE. On the one hand, a non-linearly decreasing inertia weight with the number of evolutionary iterations is employed in the WCSDE to improve the update method of the bird’s nest position, enhance the balance between the exploration and development capabilities, and strengthen the local optimization capability. On the other hand, the mutation and cross-selection mechanisms of the differential evolution (DE) algorithm are introduced to make up for the lack of the mutual relationship between the populations, avoid the loss of practical information, and increase the convergence accuracy. In the experiment part, 13 classic benchmark functions are selected to execute the function optimization tasks among the standard CS, the WCSDE, and other four CS variants to verify the effectiveness of the proposed algorithm from two aspects. The results and corresponding statistical analysis reveal that the proposed algorithm has better global search ability and strengthener robustness.

Compared to the linear activation function, a suitable nonlinear activation function can accelerate the convergence speed. Based on this finding, we propose two modified Zhang neural network (ZNN) models using different nonlinear activation functions to tackle the complex-valued systems of linear equation (CVSLE) problems in this paper. To fulfill this goal, we first propose a novel neural network called NRNN-SBP model by introducing the sign-bi-power activation function. Then, we propose another novel neural network called NRNN-IRN model by introducing the tunable activation function. Finally, simulative results demonstrate that the convergence speed of NRNN-SBP and the NRNN-IRN is faster than that of the FTRNN model. On the other hand, these results also reveal that different nonlinear activation function will have a different effect on the convergence rate for different CVSLE problems.

Recently, it is difficult to simulate, analyze and control a real knowledge-based system using the correspondence Petri net (PN) when there exist many current states. To overcome the state explosion problem of PN, an efficient decomposition algorithm is presented to divide a large-scale PN into a series of corresponding sub-PNs by keeping the consistency of dynamic properties. In this novel decomposition approach, an index function is defined to judge the subnet needs to be decomposed or not. Furthermore, an exhaustive analysis on the consistency of related dynamic properties is also discussed between the original PN and the corresponding sub-PNs. Finally, a case study is carried out to illustrate the feasibility and validity of the proposed approach.

In order to reduce the storage capacity of fonts and improve the display quality of characters, this study proposes an approach to generate square Hmong characters using intelligent derivation mechanism. In the proposed approach, the process of generating square Hmong characters is transformed into the process of solving the bounding box of components using binary tree. Thus, various square Hmong characters can be generated by repeatedly performing topological transformation of limited components. A corresponding intelligent derivation model is developed to get square Hmong characters based on the glyph data of objective components obtained by intelligent derivation. Meanwhile, deterministic finite automata is also used to represent the square of square Hmong characters. Testing results of the editing software developed based on this model demonstrate that the square Hmong characters generated by the proposed intelligent derivation mechanism can meet the needs of practical applications.

Complex-valued time-varying Sylvester equation (CVTVSE) has been successfully applied into mathematics and control domain. However, the computation load of solving CVTVSE will rise significantly with the increase of sampling rate, and it is a challenging job to tackle the CVTVSE online. In this paper, a new sign-multi-power activation function is designed. Based on this new activation function, an improved complex-valued Zhang neural network (ICZNN) model for tackling the CVTVSE is established. Furthermore, the strict proof for the maximum time of global convergence of the ICZNN is given in detail. A total of two numerical experiments are employed to verify the performance of the proposed ICZNN model, and the results show that, as compared with the previous Zhang neural network (ZNN) models with different nonlinear activation functions, this ICZNN model with the sign-multi-power activation function has a faster convergence speed to tackle the CVTVSE.

To increase the convergence speed for tackling the complex-valued system of linear equation (CVSLE) problem online, a novel fully complex-valued gradient neural network (FCGNN) model is proposed. Firstly, to tackle the CVSLE problem in the complex number field instead of the real number field, the negative gradient descent information and a new nonlinear activation function are used to design a FCGNN model. Secondly, the proof for the global convergence property of the FCGNN model is deduced from the strict derivation. Finally, the simulation results show that this FCGNN model has a higher convergence rate than the original gradient neural network (GNN) model and the improved Zhang neural network (IZNN) model.

For traditional human pose estimation models rely on a large amount of human body feature information, this paper proposes an optimization model using genetic algorithm to solve the problem of multiple person body part assembly. Different from other human body parts assembly method. The method proposed in this paper depends on the joints position information, namely the sum of the connection distances between the joints as the objective function, and finds the optimal value to obtain the best human pose assembly information. The simulation results show that compared with the traditional OpenPose model, the model proposed in this paper can obtain the same human skeleton using less position information.

Breast Cancer (BC) is the most common malignant tumor for women in the world. Histopathological examination serves as basis for breast cancer diagnosis. Due to the low accuracy of histopathological images through manual judgment, the classification of histopathological images of breast cancer has become a research hotspot in the field of medical image processing. Accurate classification of images can help doctors to properly diagnoses and improve the survival rate of patients. This paper reviews the existing works on histopathological image classification of breast cancer and analysis the advantages and disadvantages of related algorithms. Findings of the histopathological image classification of the Breast Cancer study are drawn, and the possible future directions are also discussed.

A modified Cuckoo search algorithm (MCS) is proposed in this paper to improve the accuracy of the algorithm’s convergence by implementing random operators and adapt the adjustment mechanism of the Levy Flight search step length. Comparative experiments reveal that MCS can effectively adjust the search mechanism in the high-dimensional function optimization and converge to the optimal global value.

The simulation of knowledge-based systems (KBSs) has become a significant challenge owing to the rapid increase in the scale of accumulated data. The extended formalisms that are widely used to test, model, and analyze such systems include the fuzzy production rule (FPR) and fuzzy Petri net (FPN). However, with the growth in magnitude of KBSs, it has become difficult to manually generate an FPN. Hence, the authors propose an equivalent transformation algorithm that automatically models an FPN for a sizeable KBS. The proposed method produces a final FPR by initially investigating the inner-inference path(s) between FPRs, followed by a four-phase transformation algorithm that automatically generates an equivalent FPN model for the corresponding KBS rooted in the inner-inference path(s) obtained. A KBS with 13 FPRs is used to demonstrate both the validity and feasibly of the proposed transformation algorithm. The results validate the capability of the generated FPN to fully represent the complete information base contained in the corresponding KBS.

In this paper, a new method of designing RF ring oscillator using current-mode technology is presented. The current-mode low frequency oscillator theory is transferred to facilitate the design of RF ring oscillator. As an example, a current-mode RF quadrature ring oscillator (RFQRO) for $L$ -band wireless communication applications is presented. The current-mode RFQRO consists of two first-order all-pass filters, and it provides four quadrature RF sinusoidal outputs. The proposed current-mode RFQRO is verified with GlobalFoundries’ $0.18~\mu \text{m}$ 1P6M CMOS RF process. The simulated and measured results demonstrate that the proposed RFQRO works at 1.03 GHz, and its frequency tuning range is from 648.9 MHz to 1.17 GHz by adjusting the bias voltage from 0.62 to 1 V. The measured phase noise of the RFQRO is −105.5 dBc/Hz at 1 MHz offset, and it consumes 2.5 mW from 1 V supply voltage.

Fuzzy Petri net (FPN) has been extensively applied in industrial fields for knowledge-based systems or systems with uncertainty. Although the applications of FPN are known to be successful, the theoretical research of FPN is still at an initial stage. To pave a way for further study, this work explores related dynamic properties of FPN including reachability, boundedness, safeness, liveness and fairness. The whole methodology is divided into two phases. In the first phase, a comparison between elementary net system (EN_system) and FPN is established to prove that the FPN is an extensive formalism of Petri nets using a backwards-compatible extension method. Next, current research results of dynamic properties are utilized to analyze FPN model. The results illustrate that FPN model is bounded, safe, weak live and fair, and can support theoretical evidences for designing related decomposition algorithm.

Because of the self-similarity properties of nature, fractals are widely adopted as generators of natural object multimedia contents. Unfortunately, fractals are difficult to control due to their iterated function systems, and traditional researches on fractal generating visual objects focus on mathematical manipulations. In Generative Adversarial Nets (GANs), visual object generators can be automatically guided by a single image. In this work, we explore the problem of guiding fractal generators with GAN. We assume that the same category of fractal patterns is produced by a group of parameters of initial patterns, affine transformations and random noises. Connections between these fractal parameters and visual objects are modeled by a Gaussian mixture model (GMM). Generator trainings are performed as gradients on GMM instead of fractals, so that evaluation numbers of iterated function systems are minimized. The proposed model requires no mathematical expertise from the user because parameters are trained by automatic procedures of GMM and GAN. Experiments include one 2D demonstration and three 3D real-world applications, where high-resolution visual objects are generated, and a user study shows the effectiveness of artificial intelligence guidances on fractals.

Harmony search (HS) algorithm has received widely attention in various fields because of simple operation and outstanding performance. However, there exists some bottlenecks which hinders the further development of HS algorithm, such as easy trapping into local optimum and slow convergence velocity.This paper proposes an improvement HS algorithm by using segmented search strategy (SHS) to overcome these drawbacks. Comparative experiments reveal the proposed SHS algorithm has perfect performance, especially in high dimensions.

This paper proposes a new sliding mode controller, as to enhance the performance of path tracking of three-wheeled mobile robot system. To begin with, backstepping design is used to construct a new sliding mode control law for better dynamic response and robustness. Next, in order to reduce chattering, the symbolic function is replaced with the critical saturation function in the conventional reaching law. At last, the simulation results show that backstepping design combined with saturation function can not only improve the response rate and robustness of the system, but also effectively reduce the chattering.

Part-of-speech (POS) tagging is a basic problem that needs to be solved in the informationization of Chinese-Hmong mixed text including square Hmong characters, so far no one has studied it. This paper proposes a POS tagging approach for Chinese-Hmong mixed text by utilizing improved Hidden Markov Model (HMM) to expand contextual information. The results of comparative experiments based on cross-validation reveal the proposed approach has perfect performance, and is able to obtain tagging results with good consistency and high coverage.

This paper mainly describes the basic principles of extended Kalman filter and unscented Kalman filter and the application of them in target tracking based on observation distance At last the two algorithms are compared by Matlab software The simulation results show that unscented Kalman filter has better matching effect than extended Kalman filter and the former has smaller error variation and better convergence than the latter

As real systems become larger and complicated, the number of nodes of fuzzy Petri net (FPN) also increases rapidly, it further indicated that the algorithm complexity of related application using FPN grows up sharply. To solve this state-explosion issue, a decomposition algorithm using reachability tree is proposed to divide a large-scale FPN model into a series of sub-FPNs in this paper. Comparison experiments show that all obtained sub-FPNs do not destroy internal reasoning paths of the original large-scale FPN and represent a complete reasoning path from input place(s) to the output place.

Artificial bee colony algorithm (ABC) and particle swarm optimization (PSO) are both famous optimization algorithms that have been successfully applied to various optimization problems, especially in function optimization. Those two algorithms have been attracting more and more research interest because of their efficiency and simplicity. However, PSO has poor exploration capabilities and thus is easy to fall into the local optimum; Likewise, ABC has low convergence speed. To address these shortcomings, firstly, we improved the ABC with the combination of greedy selection and crossover, secondly, a sine-cosine method will be used to help PSO jump into local optimal. Finally, a new hybrid algorithm based on improved ABC and PSO are proposed. Moreover, four functions are used to verify the effectiveness of the proposed algorithm, and the results show that, compared with other well-known algorithms, ABC-PSO is more efficient, faster and more robust in function optimization.

Optimization technique are solutions for finding for specific problems and solved existing limitation on common problems. This are especially in improving the output of WEDM processing where it involves more than a single objective or multiple output to be optimized synchronously. WEDM machining process usage increased rapidly as its capability in cutting complicated design formation. Thus the precision of the work piece is required. Therefore, this paper aims to provide a quick conceptual view of optimizing WEDM process by providing fundamentals parameters involved. Henceforth, perspective on WEDM optimization process on this paper may be use for future reference.

Aiming at solving the problem of large tracking error in the tracking process of video moving targets with the unscented Kalman filtering method, a particle filter algorithm is proposed to track video moving targets to improve the tracking effect. Particle filtering is the minimum variance estimation of the system state through the posterior probability distribution. The applications of particle filtering and unscented Kalman filtering in tracking video moving target are compared by using Matlab simulation software. The results show that the particle filter has higher accuracy and better tracking performance than the unscented Kalman filter in tracking video moving target.

Fuzzy Petri net (FPN) is widely used to repre sent, model and analyse knowledge-based systems (KBSs). Meanwhile, a reachability tree is an important tool to fully represent the flow relationship of FPN and is widely applied to implement inference in industrial areas. However, the traditional reachability ignores recording the dependence relationships (&lsquo, and/or&rsquo, relationship) among the places in the neighbouring layers. This paper develops a modified reachability tree based on an and/or graph and presents a three-phase generation algorithm to model the reachability tree for the corresponding FPN automatically via fuzzy production rules (FPRs). Four cases are used to verify the correctness and feasibility of the proposed algorithm from different viewpoints, such as general FPRs, FPRs with a condition-sharing situation, FPRs with a conclusion-sharing situation, and FPRs with multi-conclusions. Simulation results reveal that the proposed approach has the ability to automatically generate the reachability tree for the corresponding FPN correctly.

Fuzzy Petri net (FPN) is a powerful tool to execute the fault diagnosis function for various industrial applications. One of the most popular approaches for fault diagnosis is to calculate the corresponding algebra forms which record flow information and three parameters of value of all places and transitions of the FPN model. However, with the rapid growth of the complexity of the real system, the scale of the corresponding FPN is also increased sharply. It indicates that the complexity of the fault diagnosis algorithm is also raised due to the increased scale of vectors and matrix. Focusing on this situation, a bidirectional adaptive fault diagnosis algorithm is presented in this article to reduce the complexity of the fault diagnosis process via removing irrelevant places and transitions of the large-scale FPN, followed by the correctness and algorithm complexity of the proposed approach that are also discussed in detail. A practical example is utilized to show the feasibility and efficacy of the proposed method. The results of the experiments illustrated that the proposed algorithm owns the ability to simplify the inference process and to reduce the algorithm complexity due to the removal of unnecessary places and transitions in the reasoning path of the appointed output place.

Analysis of the character structure characteristics can lay an information foundation for the intelligent processing of square Hmong characters. Combined with the analysis of character structure characteristics, this paper presents a definition of the linearization of square Hmong characters, a definition of equivalence class division of the structure of square Hmong characters, and proposes a decision algorithm of structure equivalence class. According to the above algorithm, the structure of square Hmong characters is divided into eight equivalent classes. Analysis of the statistical properties, including the cumulative probability distribution, complexity, and information entropy of square Hmong characters appearing in practical documents, shows that, first, more than 90% of square Hmong characters appearing in practical documents are composed of two components, and more than 80% of these characters possess a left-right, top-bottom, or lower-left-enclosed structure, second, the number of mean components in a square Hmong character is slightly greater than 2, third, the information entropy of the structure of Hmong characters is within the interval (1.19, 2.16). Results reveal that square Hmong characters appearing frequently in practical documents follow the principle of simple structure orientation.

An index function to find out if the "OR" rule is presented.An FPN is divided into a set of sub-FPNs the proposed algorithm.A theorem is provided to compute the number of inference paths present in FPN. As with Petri nets (PNs), the state space explosion has limited further studies of fuzzy Petri net (FPN), and with the rising scale of FPN, the algorithm complexity for related applications using FPN has also rapidly increased. To overcome this challenge, we propose a decomposition algorithm that includes a backwards search stage and forward strategy for further decomposition, one that divides a large-scale FPN model into a set of sub-FPN models using both a presented index function and incidence matrix. In the backward phase, according to different output places, various completed inference paths are recognized automatically. An additional decomposition operation is then executed if the "OR" rule exists for each inference path. After analysing the proposed algorithm to confirm its rigor, a proven theorem is presented that calculates the number of inference paths in any given FPN model. A case study is used to illustrate the feasibility and robust advantages of the proposed decomposition algorithm.

Artificial Bee Colony (ABC) algorithm has received wide attention due to the outstanding performance in solving complex optimization problems. However, the further application of ABC is still hindered due to some bottlenecks, such as slow convergence velocity, poor global search ability and low accuracy of solution. To overcome these problems, an improved ABC (CT-ABC) is proposed in this paper by using both a chaotic map and a tolerance based search equation (TSE). Some standard test functions are chosen to verify the performance by implementing the proposed modified ABC algorithm, basic ABC and another modified ABC (WABC). Simulation results reveal the proposed CT-ABC algorithm is able to obtain the faster convergence speeds and better accurate approximate solution.