Recently, MOEA/D (multi-objective evolutionary algorithm based on decomposition) has achieved great success in the field of multi-objective optimization and attracted a lot attention. It decomposes problem (MOP) into set scalar subproblems using uniformly distributed aggregation weight vectors provides an excellent general algorithmic framework optimization. Generally, uniformity can ensure diversity Pareto optimal solutions, however, it cannot work as well when target MOP complex front (PF;...

State-of-the-art multiobjective evolutionary algorithms (MOEAs) treat all the decision variables as a whole to optimize performance. Inspired by cooperative coevolution and linkage learning methods in field of single objective optimization, it is interesting decompose difficult high-dimensional problem into set simpler low-dimensional subproblems that are easier solve. However, with no prior knowledge about function, not clear how function. Moreover, use such decomposition method solve...

The first cooperative co-evolutionary algorithm (CCEA) was proposed by Potter and De Jong in 1994 since then many CCEAs have been successfully applied to solving various complex optimization problems. In applying CCEAs, the problem is decomposed into multiple subproblems, each subproblem solved with a separate subpopulation, evolved an individual evolutionary (EA). Through co-evolution of EA subpopulations, complete solution acquired assembling representative members from subpopulation....

Multiobjective evolutionary algorithms based on decomposition (MOEA/D) have attracted tremendous attention and achieved great success in the fields of optimization decision-making. MOEA/Ds work by decomposing target multiobjective problem (MOP) into multiple single-objective subproblems a set weight vectors. The are solved cooperatively an algorithm framework. Since vectors define search directions and, to certain extent, distribution final solution set, configuration is pivotal MOEA/Ds....

Tchebycheff decomposition represents one of the most widely used approaches that can convert a multiobjective optimization problem into set scalar subproblems. Nevertheless, geometric properties subproblem objective functions in have not been explicitly studied. This paper proposes with lp-norm constraint on direction vectors which are endowed clear property. Especially, l <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> -norm is taken as...

Many-objective optimization problems (MaOPs) pose a big challenge to the traditional Pareto-based multiobjective evolutionary algorithms (MOEAs). As number of objectives increases, mutually nondominated solutions explodes and MOEAs become invalid due loss selection pressure. Indicator-based many-objective (MaOEAs) have been proposed address this issue by enhancing environmental selection. MaOEAs are easy implement good versatility, however, they unlikely maintain population diversity...

Multitasking optimization can achieve better performance than traditional single-tasking by leveraging knowledge transfer between tasks. However, the current multitasking algorithms suffer from some deficiencies. Particularly, on high similar problems, existing might fail to take full advantage of accelerate convergence search, or easily get trapped in local optima. Whereas, low they tend negative transfer, resulting degradation. To solve these issues, this article proposes an evolutionary...

In recent years, dynamic multiobjective optimization problems (DMOPs) have drawn increasing interest. Many evolutionary algorithms (DMOEAs) been put forward to solve DMOPs mainly by incorporating diversity introduction or prediction approaches with conventional algorithms. Maintaining a good balance of population and convergence is critical the performance DMOEAs. To address above issue, DMOEA based on decision variable classification (DMOEA-DVC) proposed in this article. DMOEA-DVC divides...

Balancing population diversity and convergence is critical for evolutionary algorithms to solve many-objective optimization problems (MaOPs). In this paper, a two-round environmental selection strategy proposed pursue good tradeoff between (MaOEAs). Particularly, in the first round, solutions with small neighborhood density are picked out form candidate pool, where of solution calculated based on novel adaptive position transformation strategy. second best terms selected from pool inserted...

Evolutionary multitasking (EMT) is an emerging research direction in the field of evolutionary computation. EMT solves multiple optimization tasks simultaneously using algorithms with aim to improve solution for each task via intertask knowledge transfer. The effectiveness transfer key success EMT. multifactorial algorithm (MFEA) represents one most widely used implementation paradigms However, it tends suffer from noneffective or even negative To address this issue and performance MFEA, we...

The divide-and-conquer strategy has been widely used in cooperative co-evolutionary algorithms to deal with large-scale global optimization problems, where a target problem is decomposed into set of lower-dimensional and tractable subproblems reduce the complexity. However, such usually demands large number function evaluations obtain an accurate variable grouping. To address this issue, merged differential grouping (MDG) method proposed article based on subset–subset interaction binary...

Different from conventional single-task optimization, the recently proposed multitasking optimization (MTO) simultaneously deals with multiple tasks different types of decision variables. MTO explores underlying similarity and complementarity among component to improve process. The well-known multifactorial evolutionary algorithm (MFEA) has been successfully introduced solve problems based on transfer learning. However, it uses a simple random inter-task learning strategy, thereby resulting...

Multiobjectivization has emerged as a new promising paradigm to solve single-objective optimization problems (SOPs) in evolutionary computation, where an SOP is transformed into multiobjective problem (MOP) and solved by algorithm find the optimal solutions of original SOP. The transformation MOP can be done adding helper-objective(s) objective, decomposing objective multiple subobjectives, or aggregating subobjectives scalar objectives. bridges gap between SOPs MOPs transforming counterpart...

Human mesh recovery from one single image has achieved rapid progress recently, but many methods suffer the appearance overfitting since training data are collected along with accurate 3D annotations in controlled settings of monotonous backgrounds or simple clothes. Some regress human vertices poses to tackle above problem. However topologies have not been well exploited, and artifacts often generated. In this paper, we aim find an efficient low-cost solution reconstruction. To end, propose...

Microarray technology allows biologists to monitor expression levels of thousands genes among various tumor tissues. Identifying relevant for sample classification types is beneficial clinical studies. One the most widely used strategies multiclass data One-Versus-All (OVA) schema that divides original problem into multiple binary one class against rest. Nevertheless, microarray tend suffer from imbalanced distribution between majority and minority classes, which inevitably deteriorates...

Recently, multi-tasking optimization (MTO) has become a rising research topic in the field of evolutionary computation that attracted increasing attention academia. Comparing with single-objective (SOO) and multi-objective (MOO), MTO can solve different tasks simultaneously by utilizing inter-task similarities complementarities. Based on crossover operator, classical multifactorial algorithm (MFEA) transfers knowledge. To broaden search region accelerate convergence, this paper integrates...