Search

This study explores crop yield forecasting through weight agnostic neural networks (WANN) optimized by a modified metaheuristic. WANNs offer the potential for lighter networks with shared weights, utilizing a two-layer cooperative framework to optimize network architecture and shared weights. The proposed metaheuristic is tested on real-world crop datasets and benchmarked against state-of-the-art algorithms using standard regression metrics. While not claiming WANN as the definitive solution, the model demonstrates significant potential in crop forecasting with lightweight architectures. The optimized WANN models achieve a mean absolute error (MAE) of 0.017698 and an R-squared ( R 2 ) score of 0.886555, indicating promising forecasting performance. Statistical analysis and Simulator for Autonomy and Generality Evaluation (SAGE) validate the improvement significance and feature importance of the proposed approach. [ABSTRACT FROM AUTHOR]

We study the recent metaheuristic search algorithm for the multidimensional assignment problem (MAP) using fitness landscape theory. The analyzed algorithm performs a very large-scale neighborhood search on a set of feasible solutions to the problem. We derive properties of the landscape graphs that represent these very large-scale search algorithms acting on the solutions of the MAP. In particular, we show that the search graph is a generalization of a hypercube. We extend and generalize the original very large-scale neighborhood search to develop the variable neighborhood search. The new search is capable of searching even larger large-scale neighborhoods. We perform numerical analyses of the search graph structures for various problem instances of the MAP and different neighborhood structures of the MAP algorithm based on a very large-scale search. We also investigate the correlation between fitness (i.e., objective values) and distance (i.e., path lengths) of the local minima (i.e., sinks of the landscape). Our results can be used to design improved search-based metaheuristics for the MAP. [ABSTRACT FROM AUTHOR]

A population-based metaheuristic algorithm that takes its cues from the foraging strategy of sparrows is called the sparrow search algorithm (SSA). While SSA is competitive when compared to other algorithms, it nevertheless has a propensity to carry out imbalanced exploitation and exploration and find the local optimum. Therefore, the modified adaptive sparrow search algorithm (MASSA), an SSA modification, is created to address these problems. To increase population variety, the MASSA uses a chaotic reverse learning technique. Second, to balance the exploitation and exploration capacities, a dynamic adaptive weight is added. In the end, an adaptive spiral search technique improves algorithm performance. Among 23 classical test functions, of which 13 are multidimensional and the other 10 are fixed dimensional, the best chaotic operator is found. It is proven that MASSA is superior. Simulation studies demonstrate that the MASSA described in this study is superior to previous algorithms in terms of stability, convergence speed, and convergence accuracy. Finally, a sample robot path planning problem is resolved using MASSA, and the experimental outcomes confirmed the viability and usefulness of MASSA. [ABSTRACT FROM AUTHOR]

In a variety of image processing applications, multilevel thresholding image segmentation has gotten a lot of interest. When traditional approaches are utilised, however, the process of obtaining the ideal threshold values takes time. Despite the fact that Hybrid metaheuristic methods can be used to overcome these limits, such approaches may be ineffective when dealing with a local solution. The present study proposes a multi-level image thresholding based hybridization strategy based Sine-Cosine Crow Search Algorithm(SCCSA) to make more efficient image segmentation. The main limitation of the classical Crow Search Algorithm (CSA) is that search agents sometimes do not produce the best solutions. To update a solution to the best solution, each search agent can use Sine-Cosine Algorithm (SCA) movements to update its position accordingly. This ensures a good balance between two goals (exploration and exploitation) would improve the efficiency of the search algorithm. The optimal threshold values are searched by the chosen objective functions of the otsu's and kapur's entropy approaches. The hybrid algorithm is evaluated in 12 standard image sets and then compared with the performance of other state-of-the-art algorithms such as ICSA, SCA, CSA and ABC. Experimental results showed that, in different metrics of the output such as objective function values, PSNR, STD values, Mean, SSIM, FSIM and CPU time, the proposed algorithm is consistently higher than other algorithms. In addition, the wilcoxon test is performed using the proposed algorithm to detect the significant differences between the other algorithms. The findings indicated that the proposed SCCSA succeeds with other well-known algorithms and has dominance over robust, accurate and convergent values. [ABSTRACT FROM AUTHOR]

Community detection is a prominent research topic in Complex Network Analysis, and it constitutes an important research field on all those areas where complex networks represent a powerful interpretation tool for describing and understanding systems involved in neuroscience, biology, social science, economy, and many others. A challenging approach to uncover the community structure in complex network, and then revealing the internal organization of nodes, is Modularity optimization. In this research paper, we present an immune optimization algorithm (opt-IA) developed to detect community structures, with the main aim to maximize the modularity produced by the discovered communities. In order to assess the performance of opt-IA, we compared it with an overall of 20 heuristics and metaheuristics, among which one Hyper-Heuristic method, using social and biological complex networks as data set. Unlike these algorithms, opt-IA is entirely based on a fully random search process, which in turn is combined with purely stochastic operators. According to the obtained outcomes, opt-IA shows strictly better performances than almost all heuristics and metaheuristics to which it was compared; whilst it turns out to be comparable with the Hyper-Heuristic method. Overall, it can be claimed that opt-IA, even if driven by a purely random process, proves to be reliable and with efficient performance. Furthermore, to prove the latter claim, a sensitivity analysis of the functionality was conducted, using the classic metrics NMI, ARI and NVI. [ABSTRACT FROM AUTHOR]

The Fixed Set Search (FSS) is a novel metaheuristic that adds a learning mechanism to the Greedy Randomized Adaptive Search Procedure (GRASP). In recent publications, its efficiency has been shown on different types of combinatorial optimization problems like routing, machine scheduling and covering. In this paper the FSS is adapted to multi-objective problems for finding Pareto Front approximations. This adaptation is illustrated for the bi-objective Minimum Weighted Vertex Cover Problem (MWVCP). In this work, a simple and effective bi-objective GRASP algorithm for the MWVCP is developed in the first stage. One important characteristic of the proposed GRASP is that it avoids the use of weighted sums of objective functions in the local search and the greedy algorithm. In the second stage, the bi-objective GRASP is extended to the FSS by adding a learning mechanism adapted to multi-objective problems. The conducted computational experiments show that the proposed FSS and GRASP algorithm significantly outperforms existing methods for the bi-objective MWVCP. To fully evaluate the learning mechanism of the FSS, it is compared to the underlying GRASP algorithm on a wide range of performance indicators related to convergence, distribution, spread and cardinality. [ABSTRACT FROM AUTHOR]

Surrogate-based optimization, nature-inspired metaheuristics, and hybrid combinations have become state of the art in algorithm design for solving real-world optimization problems. Still, it is difficult for practitioners to get an overview that explains their advantages in comparison to a large number of available methods in the scope of optimization. Available taxonomies lack the embedding of current approaches in the larger context of this broad field. This article presents a taxonomy of the field, which explores and matches algorithm strategies by extracting similarities and differences in their search strategies. A particular focus lies on algorithms using surrogates, nature-inspired designs, and those created by automatic algorithm generation. The extracted features of algorithms, their main concepts, and search operators, allow us to create a set of classification indicators to distinguish between a small number of classes. The features allow a deeper understanding of components of the search strategies and further indicate the close connections between the different algorithm designs. We present intuitive analogies to explain the basic principles of the search algorithms, particularly useful for novices in this research field. Furthermore, this taxonomy allows recommendations for the applicability of the corresponding algorithms. [ABSTRACT FROM AUTHOR]

This paper presents an efficient decoding method (namely SORTED decoding) for the de facto encoding in size optimization of trusses with automatic member grouping. The efficiency of the proposed method is evaluated through numerical experiments driven by two metaheuristic algorithms of different search mechanisms: Colliding Bodies Optimization and Jaya Algorithm. The first group of experiments show that the SORTED decoding method significantly outperforms the de facto and the two other decoding methods introduced in this study (ACCUMULATED and SORTED-ACCUMULATED) in terms of solution quality; moreover, the authors proved that the superiority of the SORTED decoding is not algorithm-dependent. Considerably better member-grouping configurations that provide up to 15% material economy are discovered for some of the well-known pre-grouped benchmark problems in the second group of experiments; surprisingly, the worst member-grouping configuration discovered by the SORTED decoding for one of the problems is better than that of the pre-grouped version. [ABSTRACT FROM AUTHOR]

Mobile ad hoc networks (MANETs) comprise a collection of independent, compact-sized, and inexpensive sensor nodes, which are commonly used to sense the physical parameters in geographical locations and transmit them to base stations (BSs). Since clustering and routing are considered commonly used energy efficient techniques, several metaheuristic algorithms have been employed to determine optimal cluster heads (CHs) and routes to destinations. However, most metaheuristic techniques have failed to achieve effective clustering and routing solutions in a large search space, and the probability of generating optimal solutions is also considerably reduced. To resolve these issues, this paper presents a new metaheuristic quantum worm swarm optimization-based clustering with a secure routing protocol for MANETs, named QGSOC-SRP. The presented QGSOC-SRP technique follows a two-stage process, namely optimal CH selection and route selection. First, the QGSO algorithm derives a fitness function using four variables, the energy, distance, node degree, and trust factor, for the optimal selection of secure CHs. Second, the SRP using the oppositional gravitational search algorithm (OGSA) is applied for the optimal selection of routes to the BS. The traditional GSA is inspired by the law of gravity and interaction among masses. To improve the effectiveness of the GSA, the OGSA is derived based on the opposition-based learning concept for population initialization and generation jumping. To validate the results regarding the effectiveness of the presented OGSOC-SRP technique, a set of experiments was performed, and the results were determined using distinct measures such as the energy consumption, network lifetime, throughput, and end-to-end delay. [ABSTRACT FROM AUTHOR]

In this paper we present a hybrid metaheuristic approach called PVS (Progress and Verify Strategy) for the two-dimensional strip packing problem (2SPP). PVS relies on two procedures: a local search algorithm that delivers satisfying placements of the items on the horizontal axis, and an exact procedure that searches for the positions of the items on the vertical axis. This last one explores all the possibilities, starting with the most promising ones, and can be stopped at any moment. PVS follows a specific anytime strategy which continuously improves the current solution until it is provably optimal or a given time limit is reached. Experimental results show that the method is competitive on moderate-sized instances compared to the best known approaches. [ABSTRACT FROM AUTHOR]

Parameter-free algorithms are of an increasing interest in applications that require out-of-the-box solving techniques. For instance, scheduling in volatile environments or control parameters optimisation often requires flexible solving approaches without the need to tune or retune the parameters. In addition, parameter-free algorithms are usually beneficial in solving previously unexplored real-world problem instances. In this paper, we propose a parameter-free local search strategy to solve graph colouring, which is the underlying problem for a number of scheduling applications. Two variants of parameter-free local search are computationally investigated: PFLS-A and PFLS-B. Both of these algorithms are based on accepting strictly improving moves and operating as a random walk if no strictly improving move is found. PFLS-A and PFLS-B differ in randomised versus systematic exploration of the neighbourhood of the current solution. Their cooperative variant CPFLS is also proposed. We compare the results of these three algorithms to the results obtained by 13 other local search methods from the literature. CPFLS provides better or equal results as the other local search algorithms in 67.25 % of per-instance comparisons. This is without employing any explicit or implicit parameters that are usually used in metaheuristics. This hints the promising nature of parameter-free metaheuristics not only for this problem but also metaheuristics in general. [ABSTRACT FROM AUTHOR]

A major issue while conceiving or parameterizing an optimization heuristic is to ensure an appropriate balance between exploitation and exploration of the search. Evolution strategies and neighborhood-based metaheuristics constitute relevant high-level frameworks, which ease the problem solving but are often complex to configure. Moreover, their effective behavior, according to the particularities of the search landscapes, remains difficult to grasp. In this paper, we deeply investigate the sampled walk search algorithm, which is a local search equivalent of the (1 , λ) -evolution strategy, considering that the neighborhood relation describes mutation possibilities. We specifically designed experiments to better understand the behavior of such a strategy offering a fine way to deal with the exploration versus exploitation dilemma. The main contribution is the analysis of search trajectories by evaluating and visualizing both their width (exploration) and their height (exploitation). More generally, we aim at bringing insights about the behavior of the (1 , λ) -ES in a discrete optimization context and within a fitness landscape perspective. [ABSTRACT FROM AUTHOR]

In complexity theory, there is a widely studied grouping of optimization problems that belongs to the non-deterministic polynomial-time hard set. One of them is the set covering problem, known as one of Karp's 21 NP -complete problems, and it consists of finding a subset of decision variables for satisfying a set of constraints at the minimum feasible cost. However, due to the nature of the problem, this cannot be solved using traditional complete algorithms for hard instances. In this work, we present an improved binary version of the monkey search algorithm for solving the set covering problem. Originally, this approximate method was naturally inspired by the cognitive behavior of monkeys for climbing mountains. We propose a new climbing process with a better exploratory capability and a new cooperation procedure to reduce the number of unfeasible solutions. For testing this approach, we present a detailed computational results section, where we illustrate how this variation of the monkey search algorithm is capable of reaching various global optimums for a well-known instance set from the Beasley's OR-Library and how it outperforms many other heuristics and meta-heuristics addressed in the literature. Moreover, we add a complete statistical analysis to show the effectiveness of the proposed approach with respect to the original version. [ABSTRACT FROM AUTHOR]

Nature-inspired algorithms usually use some form of attraction and diffusion as a mechanism for exploitation and exploration. In this paper, we investigate the role of attraction and diffusion in algorithms and their ways in controlling the behavior and performance of nature-inspired algorithms. We highlight different ways of the implementations of attraction in algorithms such as the firefly algorithm, charged system search, and the gravitational search algorithm. We also analyze diffusion mechanisms such as random walks for exploration in algorithms. It is clear that attraction can be an effective way for enhancing exploitation, while diffusion is a common way for exploration. Furthermore, we also discuss the role of parameter tuning and parameter control in modern metaheuristic algorithms and then point out some key topics for further research. [ABSTRACT FROM AUTHOR]

This article considers the heterogeneous fleet vehicle routing problem, as a variant of a well-known transportation problem: the vehicle routing problem. In order to solve this particular routing problem, a variable neighborhood search with a threshold accepting mechanism is developed and implemented. The performance of the algorithm was compared to other algorithms and tested on datasets from the available literature. Computational results show that our proposed algorithm is competitive and generates new best solutions. [ABSTRACT FROM AUTHOR]

Hill climbing method is an optimization technique that is able to build a search trajectory in the search space until reaching the local optima. It only accepts the uphill movement which leads it to easily get stuck in local optima. Several extensions to hill climbing have been proposed to overcome such problem such as Simulated Annealing, Tabu Search. In this paper, an extension version of hill climbing method has been proposed and called $$\beta$$ -hill climbing. A stochastic operator called $$\beta$$ -operator is utilized in hill climbing to control the balance between the exploration and exploitation during the search. The proposed method has been evaluated using IEEE-CEC2005 global optimization functions. The results show that the proposed method is a very efficient enhancement to the hill climbing providing powerful results when it compares with other advanced methods using the same global optimization functions. [ABSTRACT FROM AUTHOR]

Quantile regression is an increasingly important topic in statistical analysis. However, fitting censored quantile regression is hard to solve numerically because the objective function to be minimized is not convex nor concave in regressors. Performance of standard methods is not satisfactory, particularly if a high degree of censoring is present. The usual approach is to simplify (linearize) estimator function, and to show theoretically that such approximation converges to optimal values. In this paper, we suggest a new approach, to solve optimization problem (nonlinear, nonconvex, and nondifferentiable) directly. Our method is based on variable neighborhood search approach, a recent successful technique for solving global optimization problems. The presented results indicate that our method can improve quality of censored quantizing regressors estimator considerably. [ABSTRACT FROM AUTHOR]

This paper presents extensive computational experiments to compare 10 heuristics and 20 metaheuristics for the maximum diversity problem (MDP). This problem consists of selecting a subset of maximum diversity from a given set of elements. It arises in a wide range of real-world settings and we can find a large number of studies, in which heuristic and metaheuristic methods are proposed. However, probably due to the fact that this problem has been referenced under different names, we have only found limited comparisons with a few methods on some sets of instances. This paper reviews all the heuristics and metaheuristics for finding near-optimal solutions for the MDP. We present the new benchmark library MDPLIB, which includes most instances previously used for this problem, as well as new ones, giving a total of 315. We also present an exhaustive computational comparison of the 30 methods on the MDPLIB. Non-parametric statistical tests are reported in our study to draw significant conclusions. [ABSTRACT FROM AUTHOR]

In recent years many algorithms have been proposed in the literature for solving the Max-Cut problem. In this paper we report on the application of a new Tabu Search algorithm to large scale Max-cut test problems. Our method provides best known solutions for many well-known test problems of size up to 10,000 variables, although it is designed for the general unconstrained quadratic binary program (UBQP), and is not specialized in any way for the Max-Cut problem. [ABSTRACT FROM AUTHOR]

This paper studies a robust graph coloring problem, which is a variant of the classical graph coloring problem, where penalties are charged for non-adjacent vertices that are assigned the same color. The problem can be formulated as an unconstrained quadratic programming problem, and has many applications in industry. Since the problem is known to be strongly NP-complete, we develop a number of metaheuristics for it, which are based on various encoding schemes, neighborhood structures, and search algorithms. Extensive experiments suggest that our metaheuristics with a partition based encoding scheme and an improvement graph based neighborhood outperform other methods tested in our study. [ABSTRACT FROM AUTHOR]