Analysis of evolutionary multi-tasking as an island model

A multifactorial evolutionary algorithm (MFEA) is a recently proposed algorithm for evolutionary multitasking, which optimizes multiple optimization tasks simultaneously. With the design of knowledge transfer among different tasks, MFEA has demonstrated the capability to outperform its single-task counterpart in terms of both convergence speed and solution quality. In MFEA, the knowledge transfer across tasks is realized via the crossover between solutions that possess different skill factors. This crossover is thus essential to the performance of MFEA. However, we note that the present MFEA and most of its existing variants only employ a single crossover for knowledge transfer, and fix it throughout the evolutionary search process. As different crossover operators have a unique bias in generating offspring, the appropriate configuration of crossover for knowledge transfer in MFEA is necessary toward robust search performance, for solving different problems. Nevertheless, to the best of our knowledge, there is no effort being conducted on the adaptive configuration of crossovers in MFEA for knowledge transfer, and this article thus presents an attempt to fill this gap. In particular, here, we first investigate how different types of crossover affect the knowledge transfer in MFEA on both single-objective (SO) and multiobjective (MO) continuous optimization problems. Furthermore, toward robust and efficient multitask optimization performance, we propose a new MFEA with adaptive knowledge transfer (MFEA-AKT), in which the crossover operator employed for knowledge transfer is self-adapted based on the information collected along the evolutionary search process. To verify the effectiveness of the proposed method, comprehensive empirical studies on both SO and MO multitask benchmarks have been conducted. The experimental results show that the proposed MFEA-AKT is able to identify the appropriate knowledge transfer crossover for different optimization problems and even at different optimization stages along the search, which thus leads to superior or competitive performances when compared to the MFEAs with fixed knowledge transfer crossover operators.

Recently, multi-tasking optimization (MTO) has become a rising research topic in the field of evolutionary computation that has attracted increasing attention of academia. Comparing with single-objective optimization (SOO) and multi-objective optimization (MOO), MTO can solve different optimization tasks simultaneously by utilizing inter-task similarities and complementarities. Based on crossover operator, the classical multifactorial evolutionary algorithm (MFEA) transfers inter-task knowledge. To broaden the search region and accelerate the convergence, this paper integrates differential evolution (DE) and opposition-based learning (OBL) into MFEA and hence proposes MFEA/DE-OBL. The motivation of integrating DE and OBL is that they have different search neighborhoods and strong complementarity with simulated binary crossover (SBX) used in MFEA. Furthermore, integrating DE and OBL can help MFEA jump out of local optima. The effectiveness and efficiency of integrating DE and OBL into MFEA are experimentally studied on a set of benchmark problems with different degrees of similarities. Experimental results demonstrate that the proposed MFEA/DE-OBL dramatically improves the performance compared with the MFEA.

Multi-tasking optimization (MTO) has attracted increasing attention in the domain of evolutionary computation. Different from single-tasking optimization, MTO can solve multiple optimization tasks simultaneously to improve the performance of solving each optimization task by inter-task knowledge transfer. Multifactorial evolutionary algorithm (MFEA) is one of the most widely used MTO algorithm based on assortative mating and vertical cultural transmission. This work extends MFEA by integrating bone route and large neighborhood local search to solve multi-objective vehicle routing problem with time window (VRPTW). The VRPTW is modeled as two related tasks, i.e., one is a multi-objective version of VRPTW (the main task), and the other is a single-objective version of VRPTW (the auxiliary task). The resultant new algorithm namely multi-objective multi-factorial memetic algorithm (MOMFMA) solve the two tasks simultaneously where the information between the tasks is exchanged in the evolutionary process. In addition to the implicit information transfer of MFEA, the bone route is introduced to enable explicit information transfer between tasks. Particularly, bone routes are constructed as semi-finished product solutions and used in large neighborhood local search. The bone route and the large neighborhood local search work together to speed up the convergence of the algorithm. MOMFMA is tested on Solomon’s 56 datasets and the experimental results demonstrate that the efficiency of MOMFMA.

In recent years, evolutionary multitasking has received attention in the evolutionary computation community. As an evolutionary multifactorial optimization method, multifactorial evolutionary algorithm (MFEA) is proposed to realize evolutionary multitasking. One concept called the skill factor is introduced to assign a preferred task for each individual in MFEA. Then, based on the skill factor, there are some multifactorial optimization solvers including swarm intelligence that have been developed. In this paper, a PSO-FA hybrid model with a model selection mechanism triggered by updating the personal best memory is applied to multifactorial optimization. The skill factor reassignment is introduced in this model to enhance the search capability of the hybrid swarm model. Then numerical experiments are carried out by using nine benchmark problems based on typical multitask situations and by comparing with a simple multifactorial PSO to show the effectiveness of the proposed method.

Evolutionary multitasking algorithms (EMT) study how to solve multiple optimization tasks simultaneously by evolutionary computation, and investigate how knowledge sharing can accelerate the convergence of individual tasks, meaning that useful knowledge gained in solving one task can be used to solve other tasks. However, as the evolutionary search continues, the learnability among tasks may decrease, leading to a decrease in the efficiency of knowledge transfer and affecting the population evolution. To solve this problem, a new multifactorial evolutionary algorithm (MFEA-VOM) is proposed in this paper, which applies to three strategies, namely, implicit conversion strategy, opposition matrix strategy, and regulatory gene fusion strategy. The implicit conversion strategy is applied to minimize the threat of negative knowledge migration and reduce the impact caused by negative knowledge migration. The proposed opposition matrix strategy explores more unknown areas of the population and improves the exploration ability of the population by further exploring and utilizing the unified search space, transforming the parent individuals into an appropriate task through mapping relationships, and reducing the gap between tasks. The proposed regulatory gene fusion strategy is applied to the reproduction of individuals to produce better individuals applicable to the task, submitting the efficiency of knowledge transfer. Through a comprehensive experimental analysis of the EMT optimization problem, the experimental results demonstrate the better performance of MFEA-VOM compared to other EMT algorithms.

Large Scale Global Optimization (LSGO) is interesting for its applications in machine learning, e.g. in deep learning or knowledge graph embeddings. Evolutionary algorithms (EA) have been found efficient in solving complex optimization problems. However, the performance of conventional EAs degrades with the increasing number of decision variables due to the lack of scalability. This paper proposes a scalable, parallel, distributed and hybrid EA named Distributed Scalable Shade-BatScalable Shade Bat (DistSSB) to solve LSGO problems. DistSSB is inspired by the exploration capability of the SHADE algorithm and exploitation feature of the Bat algorithm (BA). To achieve scalability, DistSSB is implemented using the popular distributed in-memory framework, Apache Spark. DistSSB distributes its population into multiple sub-populations using the island model. Each sub-population is independently evolved using SHADE or BA. After the migration interval, the best solutions are broadcasted employing the mesh topology. We have compared the scalability, efficiency, and efficacy of DistSSB with SHADE-ILS (CEC-2018 Winner) and GL-SHADE algorithm on the CEC-2013 LSGO benchmark function suite.For most functions, DistSSB has obtained better optimization results in lesser execution time as compared to SHADE-ILS and GL-SHADE. We have tested and shown the scalability of DistSSB for up to "one million" dimensions, whereas SHADE-ILS and GLSHADE fail to scale up for larger problems.

This paper presents novel frameworks for efficient and lightweight convolutional neural networks that are suitable for use in embedded system applications. Population-based metaheuristic approaches including the genetic algorithm, cuckoo search, the multifactorial evolutionary algorithm and a proposed hybrid approach are used to optimize their performance on image classification tasks. The methods utilize small population sizes without requiring weight-sharing or a surrogate function, and both binary and integer encoding methods are applied in the optimization. The results from these various strategies are evaluated using metrics of computation time and classification accuracy. The multifactorial-based approach is found to give the highest classification accuracy and it requires only a moderate evaluation time. Also, comparisons with prior approaches demonstrate that these methods show a favorable tradeoff between accuracy and computational cost.

In recent years, one of the prominent research areas in the complex network field has been the Influence Maximization Problem. This problem focuses on selecting seed sets to achieve optimal information propagation across networks. Practical networks often encounter challenges like node or link failures due to internal issues or external disturbances. Addressing this, researchers emphasize seed robustness against potential interferences, framing it as the robust influence maximisation problem. However, current approaches to this problem are incomplete, leaving several challenges unaddressed. On one hand, existing methods primarily handle seed selection under isolated disruptions, neglecting the simultaneous threats posed by both node-based and link-based attacks. On the other hand, prevailing algorithms fail to capture information synergy from multiple scenarios during the solution process. To bridge these gaps, this study integrates the multi-tasking optimisation theory into robust influence maximisation, introducing an evolutionary algorithm called DMFEA. DMFEA concurrently addresses multiple optimization scenarios, leveraging synergy between tasks while emphasizing information diversity. Experimental results demonstrate DMFEA's competitive edge over existing methods. This research significantly advances collaborative optimization for robust influence maximisation under multi-scenario disruptions, offering a reliable solution for robust information diffusion in complex environments.

Evolutionary algorithms (EAs) have shown their great capability of handling optimization problems. In the domain of large scale global optimization, many distributed EAs (dEAs) have been proposed for maintaining population diversity so as to enhance their efficacy. One well-known variant of dEA is the island model EA, in which several populations form islands communicating through migration. There are several key factors that affect the performance and search behavior of island model EA, such as population size of each island and migration topology. While most studies of dEA focus on low or medium dimensional problems, an investigation into the effects of these factors on high dimensional problems is greatly needed. This study presents an empirical analysis of island model EA on large scale global optimization problems. The analysis examines the solution quality, convergence speed, and population diversity of island model EA with different migration topologies, population sizes, migration rates, and migration frequencies on four benchmark function of 1,000 dimensions. The results render guidelines for using island model EA to solve large scale global optimization problems.

Universal strategy of dynamic subpopulation number management in practical network optimization problems

Island Models are parallel approaches to Evolutionary Algorithms that not only offer the benefits of parallelization but are also regarded as models with an extensively distinct behaviour. This study applies for the first time an Island Model to the optimization of short-ranged Morse clusters, combined with a hybrid steady-state evolutionary algorithm and a local optimization method. Different migration parameters are experimented and the resulting behaviours are extensively analysed. Results are compared to a state-of-the-art sequential approach, showing slight improvements. Differences in behaviour between the Island Model and the sequential approach are comprehensively discussed. This study shows that Island Models are a competitive parallel approach with promising results on cluster geometry optimization problems.

SummaryEvolutionary algorithms are one of the most popular forms of optimization algorithms. They are comparatively easy to use and were successfully employed for a wide variety of practical applications. However, frequently, it is necessary to execute them in parallel in order to reduce the runtime. There are a number of different approaches for the parallelization of evolutionary algorithms, and various hardware platforms can be used for the parallel execution. However, not every platform is equally suitable for any kind of parallelization of evolutionary algorithms. In addition, it also depends on properties of the concrete optimization problem to be solved and on the used evolutionary algorithm, which platform is best suited for the execution. The present work observes this in detail for two common forms of parallelization of evolutionary algorithms – the island model and the global parallelization – and for four widely used parallel computing platforms – multi‐core CPUs, clusters, graphics cards, and grids. Based on empirical and analytical investigations, it is determined, under which circumstances an architecture is better suited for the execution of a parallel evolutionary algorithm than another (andvice versa). Guidelines are derived that support users of parallel evolutionary algorithms with the choice of an appropriate platform. Copyright © 2016 John Wiley & Sons, Ltd.

We propose a parallel distributed model of a hybrid fuzzy genetics-based machine learning (GBML) algorithm to drastically decrease its computation time. Our hybrid algorithm has a Pittsburgh-style GBML framework where a rule set is coded as an individual. A Michigan-style rule-generation mechanism is used as a kind of local search. Our parallel distributed model is an island model where a population of individuals is divided into multiple islands. Training data are also divided into multiple subsets. The main feature of our model is that a different training data subset is assigned to each island. The assigned training data subsets are periodically rotated over the islands. The best rule set in each island also migrates periodically. We demonstrate through computational experiments that our model decreases the computation time of the hybrid fuzzy GBML algorithm by an order or two of magnitude using seven parallel processors without severely degrading the generalization ability of obtained fuzzy rule-based classifiers. We also examine the effects of the training data rotation and the rule set migration on the search ability of our model.

In MFEA, the knowledge transfer across tasks is realized via the crossover between solutions that possess different skill factors.

In differential evolution (DE) algorithm, numerous adaptive methods based on superior individual information in the current generation have been proposed. However, the individual difference between two generations, which represents whether the corresponding parameters and mutation strategy are suitable for this individual, has not been utilized. Considering that different (superior or inferior) individuals need different parameters and strategies, a new DE variant (DI-DE), which is directed by individual difference information between generations and individual information in the current generation to obtain optimal control parameters and an offspring generation strategy, is proposed. In DI-DE, every individual possesses its own parameters and strategy. First, individuals are distinguished as superior or inferior depending on their fitness values in the current generation. The parameters are tuned in accordance with the information on superior individuals. In addition, the conception of potential individuals is proposed for superior and inferior individuals on the basis of the individual difference information between two generations. By learning from the current and past information, the suitable mutation strategy is adjusted for superior and inferior individuals on the basis of the experience of potential individuals to help them become potential individuals in the next generation. DI-DE is compared with 28 excellent algorithms on three well-known benchmark sets (CEC2005, CEC2013, and CEC2014) of low dimensionality and one large scale benchmarks set (CEC LSGO 2013). Experimental results demonstrate the competitive performance of DI-DE. Finally, DI-DE is applied to optimize the operation conditions of PX oxidation process.