Search

In computational optimization, the Beetle Antennae Search algorithm is renowned for its bio-inspired mechanics and robust performance. However, its efficacy is often challenged by complex, multimodal landscapes in real-world applications. This study introduces an innovative methodology leveraging OpenAI’s ChatGPT to enhance the Beetle Antennae Search algorithm. Using ChatGPT, we developed complex benchmark optimization functions as testing grounds for the algorithm and its iterations. Through AI-guided exploration, two advanced variants were implemented: Adaptive Beetle Antennae Search, with dynamic parameter adjustment, and Adaptive Feedback Beetle Antennae Search, integrating a feedback mechanism for self-tuning. These variants were rigorously tested against the AI-suggested benchmarks, demonstrating superior convergence and precision. For example, the Adaptive Feedback Beetle Antennae Search significantly improved results on the Rastrigin and Ackley functions. This study exemplifies the transformative role of AI in algorithmic design and development, showcasing how AI can assist in creating more efficient optimization methods. By detailing our AI-assisted approach, we contribute to expanding optimization techniques and demonstrate AI’s potential as a co-creator in scientific and engineering advancements.

Abstract In engineering fields, time‐varying matrix inversion (TVMI) issue is often encountered. Zeroing neural network (ZNN) has been extensively employed to resolve the TVMI problem. Nevertheless, the original ZNN (OZNN) and the integral‐enhanced ZNN (IEZNN) usually fail to deal with the TVMI problem under unbounded noises, such as linear noises. Therefore, a neural network model that can handle the TVMI under linear noise interference is urgently needed. This paper develops a double integral‐enhanced ZNN (DIEZNN) model based on a novel integral‐type design formula with inherent linear‐noise tolerance. Moreover, its convergence and robustness are verified by derivation strictly. For comparison and verification, the OZNN and the IEZNN models are adopted to resolve the TVMI under multiple identical noise environments. The experiments proved that the DIEZNN model has excellent advantages in solving TVMI problems under linear noises. In general, the DIEZNN model is an innovative work and is proposed for the first time. Satisfyingly, the errors of DIEZNN are always less than 1 × 10−3 under linear noises, whereas the error norms of OZNN and IEZNN models are not convergent to zero. In addition, these models are applied to the control of the controllable permanent magnet synchronous motor chaotic system to indicate the superiority of the DIEZNN.

Abstract Time‐varying matrix inversion is an important field of matrix research, and lots of research achievements have been obtained. In the process of solving time‐varying matrix inversion, disturbances inevitably exist, thus, a model that can suppress disturbance while solving the problem is required. In this paper, an advanced continuous‐time recurrent neural network (RNN) model based on a double integral RNN design formula is proposed for solving continuous time‐varying matrix inversion, which has incomparable disturbance‐suppression property. For digital hardware applications, the corresponding advanced discrete‐time RNN model is proposed based on the discretisation formulas. As a result of theoretical analysis, it is demonstrated that the advanced continuous‐time RNN model and the corresponding advanced discrete‐time RNN model have global and exponential convergence performance, and they are excellent for suppressing different disturbances. Finally, inspiring experiments, including two numerical experiments and a practical experiment, are presented to demonstrate the effectiveness and superiority of the advanced discrete‐time RNN model for solving discrete time‐varying matrix inversion with disturbance‐suppression.

In recent years, neural networks have become a common practice in academia for handling complex problems. Numerous studies have indicated that complex problems can generally be formulated as a single or a set of time-varying equations. Dynamic neural networks, as powerful tools for processing time-varying problems, play an essential role in their online solution. This paper reviews recent advances in real-valued, complex-valued, and noise-tolerant dynamic neural networks for solving various time-varying problems, discusses the finite-time convergence, fixed/varying parameters, and noise tolerance properties of dynamic neural network models. This review is highly relevant for researchers and novices interested in using dynamic neural networks to solve time-varying problems.

The rapid expansion of artificial intelligence (AI) technologies presents novel technical solutions to traditional accounting and finance problems. Despite this, scholars in accounting and finance frequently encounter difficulties navigating the extensive and intricate domain knowledge of AI and its continuously evolving literature. To address this gap, this paper conducts a qualitative survey of the implementation of AI methods in accounting and finance. The paper is structured into four sections. Firstly, we examine the conventional accounting and finance issues and their requirement for AI techniques. Secondly, to inform accounting and finance researchers about the potential of AI, we present broad categories of AI applications. Thirdly, we explore recent research on AI solutions to conventional problems. Finally, we highlight emerging trends and possible research directions.

Plant height and leaf area index (LAI) are crucial growth indicators that reflect the growth status of tomatoes in greenhouses, enabling accurate determinations to effectively estimate crop transpiration and formulate irrigation strategies for reducing agricultural water waste. There is a need for the increased application of related models to simulate tomato growth indices in the traditional greenhouse production in China. This study proposes a nondestructive, real-time monitoring and simulation device for measuring tomato plant height and leaf area index. The weight of aboveground tomatoes was obtained by suspending tomato plants on dynamometers, while the total weight of stem and leaf organs was determined using a distribution coefficient simulation model. The R2 value between the measurements from the electronic scale and those from the aboveground fresh weight device for tomatoes was 0.937, with an RMSE value of 0.05 kg. The monitoring device did not affect the average tomato growth during operation. The device will not affect the growth of tomatoes during monitoring. A multiple linear regression was used to compare the measured and simulated values of the plant height and leaf area index of various types of greenhouse tomatoes cultivated in different greenhouse types. The average R2 value for simulating plant height was 0.817 with an RMSE of 10.81 cm. The average R2 value for the leaf area index was 0.854, with an RMSE of 0.55 m2·m−2. The simulated values for plant height and leaf area index closely matched the measured values, indicating that the model has high accuracy and applicability in traditional Chinese greenhouses (solar greenhouses and insulated plastic greenhouses). However, further optimization is required for commercially produced, continuous plastic greenhouses equipped with greenhouse environmental control equipment.

The field of computer science has undergone rapid expansion due to the increasing interest in improving system performance. This has resulted in the emergence of advanced techniques, such as neural networks, intelligent systems, optimization algorithms, and optimization strategies. These innovations have created novel opportunities and challenges in various domains. This paper presents a thorough examination of three intelligent methods: neural networks, intelligent systems, and optimization algorithms and strategies. It discusses the fundamental principles and techniques employed in these fields, as well as the recent advancements and future prospects. Additionally, this paper analyzes the advantages and limitations of these intelligent approaches. Ultimately, it serves as a comprehensive summary and overview of these critical and rapidly evolving fields, offering an informative guide for novices and researchers interested in these areas.

This research proposes and investigates some improvements in gradient descent iterations that can be applied for solving system of nonlinear equations (SNE). In the available literature, such methods are termed improved gradient descent methods. We use verified advantages of various accelerated double direction and double step size gradient methods in solving single scalar equations. Our strategy is to control the speed of the convergence of gradient methods through the step size value defined using more parameters. As a result, efficient minimization schemes for solving SNE are introduced. Linear global convergence of the proposed iterative method is confirmed by theoretical analysis under standard assumptions. Numerical experiments confirm the significant computational efficiency of proposed methods compared to traditional gradient descent methods for solving SNE.

Engineering design optimization problems are difficult to solve because the objective function is often complex, with a mix of continuous and discrete design variables and various design constraints. Our research presents a novel hybrid algorithm that integrates the benefits of the sine cosine algorithm (SCA) and artificial bee colony (ABC) to address engineering design optimization problems. The SCA is a recently developed metaheuristic algorithm with many advantages, such as good search ability and reasonable execution time, but it may suffer from premature convergence. The enhanced SCA search equation is proposed to avoid this drawback and reach a preferable balance between exploitation and exploration abilities. In the proposed hybrid method, named HSCA, the SCA with improved search strategy and the ABC algorithm with two distinct search equations are run alternately during working on the same population. The ABC with multiple search equations can provide proper diversity in the population so that both algorithms complement each other to create beneficial cooperation from their merger. Certain feasibility rules are incorporated in the HSCA to steer the search towards feasible areas of the search space. The HSCA is applied to fifteen demanding engineering design problems to investigate its performance. The presented experimental results indicate that the developed method performs better than the basic SCA and ABC. The HSCA accomplishes pretty competitive results compared to other recent state-of-the-art methods.

The problem of portfolio management relates to the selection of optimal stocks, which results in a maximum return to the investor while minimizing the loss. Traditional approaches usually model the portfolio selection as a convex optimization problem and require the calculation of gradient. Note that gradient-based methods can stuck at local optimum for complex problems and the simplification of portfolio optimization to convex, and further solved using gradient-based methods, is at a high cost of solution accuracy. In this paper, we formulate a nonconvex model for the portfolio selection problem, which considers the transaction cost and cardinality constraint, thus better reflecting the decisive factor affecting the selection of portfolio in the real-world. Additionally, constraints are put into the objective function as penalty terms to enforce the restriction. Note that this reformulated problem cannot be readily solved by traditional methods based on gradient search due to its nonconvexity. Then, we apply the Beetle Antennae Search (BAS), a nature-inspired metaheuristic optimization algorithm capable of efficient global optimization, to solve the problem. We used a large real-world dataset containing historical stock prices to demonstrate the efficiency of the proposed algorithm in practical scenarios. Extensive experimental results are presented to further demonstrate the efficacy and scalability of the BAS algorithm. The comparative results are also performed using Particle Swarm Optimizer (PSO), Genetic Algorithm (GA), Pattern Search (PS), and gradient-based fmincon (interior-point search) as benchmarks. The comparison results show that the BAS algorithm is six times faster in the worst case (25 times in the best case) as compared to the rival algorithms while achieving the same level of performance.

A novel meta-heuristic algorithm named Egret Swarm Optimization Algorithm (ESOA) is proposed in this paper, which is inspired by two egret species’ hunting behavior (Great Egret and Snowy Egret). ESOA consists of three primary components: a sit-and-wait strategy, aggressive strategy as well as discriminant conditions. The learnable sit-and-wait strategy guides the egret to the most probable solution by applying a pseudo gradient estimator. The aggressive strategy uses random wandering and encirclement mechanisms to allow for optimal solution exploration. The discriminant model is utilized to balance the two strategies. The proposed approach provides a parallel framework and a strategy for parameter learning through historical information that can be adapted to most scenarios and has well stability. The performance of ESOA on 36 benchmark functions as well as 3 engineering problems are compared with Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Differential Evolution (DE), Grey Wolf Optimizer (GWO), and Harris Hawks Optimization (HHO). The result proves the superior effectiveness and robustness of ESOA. ESOA acquires the winner in all unimodal functions and reaches statistic scores all above 9.9, while the scores are better in complex functions as 10.96 and 11.92.

The recently emerging multi-portfolio selection problem lacks a proper framework to ensure that client privacy and database secrecy remain intact. Since privacy is of major concern these days, in this paper, we propose a variant of Beetle Antennae Search (BAS) known as Distributed Beetle Antennae Search (DBAS) to optimize multi-portfolio selection problems without violating the privacy of individual portfolios. DBAS is a swarm-based optimization algorithm that solely shares the gradients of portfolios among the swarm without sharing private data or portfolio stock information. DBAS is a hybrid framework, and it inherits the swarm-like nature of the Particle Swarm Optimization (PSO) algorithm with the BAS updating criteria. It ensures a robust and fast optimization of the multi-portfolio selection problem whilst keeping the privacy and secrecy of each portfolio intact. Since multi-portfolio selection problems are a recent direction for the field, no work has been done concerning the privacy of the database nor the privacy of stock information of individual portfolios. To test the robustness of DBAS, simulations were conducted consisting of four categories of multi-portfolio problems, where in each category, three portfolios were selected. To achieve this, 200 days worth of real-world stock data were utilized from 25 NASDAQ stock companies. The simulation results prove that DBAS not only ensures portfolio privacy but is also efficient and robust in selecting optimal portfolios.

Complex time-dependent Lyapunov equation (CTDLE), as an important means of stability analysis of control systems, has been extensively employed in mathematics and engineering application fields. Recursive neural networks (RNNs) have been reported as an effective method for solving CTDLE. In the previous work, zeroing neural networks (ZNNs) have been established to find the accurate solution of time-dependent Lyapunov equation (TDLE) in the noise-free conditions. However, noises are inevitable in the actual implementation process. In order to suppress the interference of various noises in practical applications, in this paper, a complex noise-resistant ZNN (CNRZNN) model is proposed and employed for the CTDLE solution. Additionally, the convergence and robustness of the CNRZNN model are analyzed and proved theoretically. For verification and comparison, three experiments and the existing noise-tolerant ZNN (NTZNN) model are introduced to investigate the effectiveness, convergence and robustness of the CNRZNN model. Compared with the NTZNN model, the CNRZNN model has more generality and stronger robustness. Specifically, the NTZNN model is a special form of the CNRZNN model, and the residual error of CNRZNN can converge rapidly and stably to order 10−5 when solving CTDLE under complex linear noises, which is much lower than order 10−1 of the NTZNN model. Analogously, under complex quadratic noises, the residual error of the CNRZNN model can converge to 2∥A∥F/ζ3 quickly and stably, while the residual error of the NTZNN model is divergent.

Deep Convolutional Neural Networks (CNNs) represent the state-of-the-art artificially intelligent computing models for image classification. The advanced cognition and pattern recognition abilities possessed by humans are ascribed to the intricate and complex neurological connection in human brains. CNNs are inspired by the neurological structure of the human brain and show performance at par with humans in image recognition and classification tasks. On the lower extreme of the neurological complexity spectrum lie small organisms such as insects and worms, with simple brain structures and limited cognition abilities, pattern recognition, and intelligent decision-making abilities. However, billions of years of evolution guided by natural selection have imparted basic survival instincts, which appear as an “intelligent behavior”. In this paper, we put forward the evidence that a simple algorithm inspired by the behavior of a beetle (an insect) can fool CNNs in image classification tasks by just perturbing a single pixel. The proposed algorithm accomplishes this in a computationally efficient manner as compared to the other adversarial attacking algorithms proposed in the literature. The novel feature of the proposed algorithm as compared to other metaheuristics approaches for fooling a neural network, is that it mimics the behavior of a single beetle and requires fewer search particles. On the contrary, other metaheuristic algorithms rely on the social or swarming behavior of the organisms, requiring a large population of search particles. We evaluated the performance of the proposed algorithm on LeNet-5 and ResNet architecture using the CIFAR-10 dataset. The results show a high success rate for the proposed algorithms. The proposed strategy raises a concern about the robustness and security aspects of artificially intelligent learning systems.

With the emergence of various online trading technologies, fraudulent cases begin to occur frequently. The problem of fraud in public trading companies is a hot topic in financial field. This paper proposes a fraud detection model for public trading companies using datasets collected from SEC’s Accounting and Auditing Enforcement Releases (AAERs). At the same time, this computational finance model is solved with a nonlinear activated Beetle Antennae Search (NABAS) algorithm, which is a variant of the meta-heuristic optimization algorithm named Beetle Antennae Search (BAS) algorithm. Firstly, the fraud detection model is transformed into an optimization problem of minimizing loss function and using the NABAS algorithm to find the optimal solution. NABAS has only one search particle and explores the space under a given gradient estimation until it is less than an “Activated Threshold” and the algorithm is efficient in computation. Then, the random under-sampling with AdaBoost (RUSBoost) algorithm is employed to comprehensively evaluate the performance of NABAS. In addition, to reflect the superiority of NABAS in the fraud detection problem, it is compared with some popular methods in recent years, such as the logistic regression model and Support Vector Machine with Financial Kernel (SVM-FK) algorithm. Finally, the experimental results show that the NABAS algorithm has higher accuracy and efficiency than other methods in the fraud detection of public datasets.

Many hard optimization problems have been efficiently solved by two notable swarm intelligence algorithms, artificial bee colony (ABC) and firefly algorithm (FA). In this paper, a collaborative hybrid algorithm based on firefly and multi-strategy artificial bee colony, abbreviated as FA-MABC, is proposed for solving single-objective optimization problems. In the proposed algorithm, FA investigates the search space globally to locate favorable regions of convergence. A novel multi-strategy ABC is employed to perform local search. The proposed algorithm incorporates a diversity measure to help in the switch criteria. The FA-MABC is tested on 40 benchmark functions with diverse complexities. Comparative results with the basic FA, ABC and other recent state-of-the-art metaheuristic algorithms demonstrate the competitive performance of the FA-MABC.

In this paper, composite films composed of soya bean protein isolate (SPI) and polyacrylamide (PAM) were prepared under variations of ultrasonic power, treatment time and heating temperature. The effects of the major processing parameters related to ultrasonic-assisted treatment were evaluated and optimized through the single-factor analysis and Box–Behnken design (BBD), respectively, when the tensile strength of composite films was considered as the response value. The single-factor analysis was carried out to study the effects of ultrasonic power, treatment time and heating temperature on the viscosity and cohesion of the slurry and the tensile strength of SPI/PAM composite films, which also provided a reasonable data range of each factor for further optimization. Experiment results indicated that these three factors play a significant role in the tensile strength of films. Then BBD was applied to optimize the treatment conditions of these three factors, using the tensile strength of films as the response value. According to the interactive second-order polynomial model of three factors and the three-dimensional response surface, the maximum tensile strength of films was obtained under the optimal condition. To verify the reliability of the model, the experiment with the optimal condition was conducted, and results demonstrated that the observed tensile strength was in agreement with the predicated one. Also, the morphology and water solubility of the films showed that the film can be coated on the yarns evenly and removed clearly.

Malnutrition in cancer patients is the result of a variety of factors. Decreased intake, absorption disorders, metabolic disorders and increased REE are the main causes of malnutrition. The incidence of malnutrition in cancer patients is higher than that in patients with benign diseases, and the consequences are more serious. Malnourished cancer patients have more comorbidities and complications, higher medical costs, shorter survival times, and decreased tolerance and sensitivity to tumor-control therapy. Therefore, cancer patients are in greater need of nutrition therapy, which should become a necessary basic treatment and a first-line treatment for cancer. NST should be the core member of the multidisciplinary integrated treatment [multidisciplinary treatment (MDT) to holistic integrative management (HIM)] of cancer. [ABSTRACT FROM AUTHOR]

In this paper, we delve into the transformative influence of Large Language Models (LLMs) in the financial sector. Through meticulous exploration, we uncover the multifaceted applications of LLMs, ranging from elevating customer support and fortifying fraud detection to reshaping market analysis and prediction. LLMs, with their unparalleled ability to process extensive textual data, bring forth innovative solutions and insights. However, we also address critical challenges such as user trust and ethical considerations, emphasizing the need for responsible integration. Collaborative efforts between industry stakeholders and researchers are essential prerequisites for making a pivotal stride towards a future where LLMs redefine financial practices, with efficiency, accuracy, and ethical precision shaping the industry's evolution. [ABSTRACT FROM AUTHOR]