NSGA-II With Simple Modification Works Well on a Wide Variety of Many-Objective Problems

Pang, Lie Meng; Ishibuchi, Hisao; Shang, Ke

doi:10.1109/access.2020.3032240

Cited by 37 publications

(10 citation statements)

References 21 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The -dominance criterion lets each solution have a larger dominating area, thereby increasing the probability of DRSs being dominated by Pareto optimal solutions. mNSGA-II: As reported in [16], the modification of the objective values of each solution can decrease the negative effect of DRSs. The modified objective value is defined as:…”

Section: Moeas For Drs Eliminationmentioning

confidence: 99%

The dilemma between eliminating dominance-resistant solutions and preserving boundary solutions of extremely convex Pareto fronts

Wang

Yang

et al. 2021

Complex Intell. Syst.

Self Cite

View full text Add to dashboard Cite

It has been acknowledged that dominance-resistant solutions (DRSs) extensively exist in the feasible region of multi-objective optimization problems. Recent studies show that DRSs can cause serious performance degradation of many multi-objective evolutionary algorithms (MOEAs). Thereafter, various strategies (e.g., the $$\epsilon $$ ϵ -dominance and the modified objective calculation) to eliminate DRSs have been proposed. However, these strategies may in turn cause algorithm inefficiency in other aspects. We argue that these coping strategies prevent the algorithm from obtaining some boundary solutions of an extremely convex Pareto front (ECPF). That is, there is a dilemma between eliminating DRSs and preserving boundary solutions of the ECPF. To illustrate such a dilemma, we propose a new multi-objective optimization test problem with the ECPF as well as DRSs. Using this test problem, we investigate the performance of six representative MOEAs in terms of boundary solutions preservation and DRS elimination. The results reveal that it is quite challenging to distinguish between DRSs and boundary solutions of the ECPF.

show abstract

Section: Moeas For Drs Eliminationmentioning

confidence: 99%

The dilemma between eliminating dominance-resistant solutions and preserving boundary solutions of extremely convex Pareto fronts

Wang

Yang

et al. 2021

Complex Intell. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Taking motivation from [57], we assume that the source data ( ) * + of population size had been archived at every iteration of an MOEA, when solving the source task of dimensionality . Different from the methodology proposed in Section 4.3.1, here we adopt the preference relationship between solutions based on the concepts of nondominated front (NF) and crowding distance (CD) [72] that are lexicographically considered in a multi-objective optimization setting; given the common knowledge of these terms in the associated literature, we refrain from providing a detailed discussion about them herein for the sake of brevity. Accordingly, a solution is preferred over another solution if any one of the following conditions holds true [79]: (i) or (ii) ; see Figure 5.1(a) for a pictorial depiction.…”

Section: Methodology For Learningmentioning

confidence: 99%

“…Multi-objective EAs (MOEAs), by virtue of their population-based search strategy, have gained popularity for solving MOPs as they are able to reasonably approximate a set of alternative solutions in the multi-objective optimization setting within a single run. Some of the most widely used MOEAs today include the Pareto dominance-based NSGA-II [71,72] and SPEA2 [73,74], the decomposition-based MOEA/D [75,76], and the preference-based PBEA [77], to name a few. Just like traditional EAs, many existing MOEAs start their search from scratch without any attempt to exploit related source knowledge from past optimization experiences.…”

Section: Problem-solving In Treomentioning

confidence: 99%

“…The proposed MOTrEO+ is detailed in Algorithm 5.1, according to the novel framework established in Section 3.4. Our algorithm assumes (without loss of generality) the popular NSGA-II [72] to be the base MOEA. Hence, the concepts of non-dominated sorting and crowding distances to evaluate the relative fitness of candidate solutions in multi-dimensional objective spaces are adopted herein.…”

Section: Algorithm Summarymentioning

confidence: 99%

“…Four optimizers are considered for comparison in this our experiments. They are: (i) the base NSGA-II [72], (ii) a multi-objective variant of the edge histogram-based sampling algorithm (EHBSA) which is a probabilistic model-based optimizer (without knowledge transfer) for combinatorial problems [119], (iii) the AMTEA without solution representation learning [10], and (iv) our proposed MOTrEO+ . All search populations of size consist of permutation-coded solutions.…”

Section: Experimental Configurationmentioning

confidence: 99%

See 2 more Smart Citations

Solution representation learning in transfer evolutionary optimization

Lim¹

View full text Add to dashboard Cite

The human cognitive ability to learn with experience is a masterpiece of natural evolution that has yet to be fully duplicated in computational and artificial intelligence systems. When presented with a new task, our brain has the natural tendency to retrieve and reuse knowledge priors acquired from related experiences, thereby speeding up our problem-solving process. In the modern era of data-driven optimization, fuelled by growing amounts of data and seamless information transmission technologies, it is becoming increasingly important for machines to embody the ability to learn from experiences as well. To this end, a recent computational paradigm known as transfer evolutionary optimization (TrEO) has emerged to encompass methods that leverage knowledge priors from various source optimization tasks to boost the convergence performance in a new but related target task. Despite the potential for performance speed-up, the effectiveness of existing TrEO algorithms in actual practice could be hampered by discrepancies between the original search spaces of source and target problemsa common scenario when solving blackbox optimization problems of diverse properties, where little is known about their search landscapes a priori. In particular, heterogeneous source and target dimensionalities or a lack of overlap between their optimized search distributions may give rise to unaligned solution representations that conceal useful inter-task relationships. This could in turn cause two undesired issues in TrEO, namely, the scarcity of beneficial positive transfers, or the increase of harmful negative transfers. Even though methodological advances have been made, for instance, the state-of-the-art probabilistic model-based transfer approach for curbing negative inter-task interactions, there is currently a lack of approaches capable of jointly addressing the two predominant issues in TrEO.Taking the cue, this thesis presents a novel study on solution representation learning in probabilistic model-based TrEO for inducing greater alignment (of solution representations) and hence positive transfers between distinct optimization tasks that bear discrepancies in their original search spaces, while simultaneously curbing the Abstract ii occurrence of negative transfers. A formalization of solution representation learning in TrEO is established. To this end, the importance and motivations of solution representation learning for uncovering useful but hidden inter-task relationships are conceived. Following from the motivations, a principled perspective of solution representation learning via search space mappings in TrEO is presented, where a streamlined definition is first given. Accordingly, the source-to-target map is proposed as a category of spatial mappings by which source tasks are mapped to a transformed and well-aligned solution representation space, such that the discrepancies between various sources and the target task are reducedin this category, the target search space serves as the transformed (or common) solution repre...

show abstract

NSGA‐II algorithm‐based automated cigarette finished goods storage level optimization research

Hu,

Dong,

Wang

et al. 2023

Adv Control Appl

View full text Add to dashboard Cite

With the growth of Internet of Things technology, more and more businesses are implementing automated cargo storage systems. By using an appropriate automated storage space allocation model, these businesses can significantly reduce their storage pressure while saving money on logistics and increasing the effectiveness of their product distribution. Therefore, the study is based on the non‐dominated sorting genetic algorithms II (non‐dominated sorting genetic algorithm, NSGA II), which combines the three basic principles of space allocation as the objective function applied to the allocation model of the algorithm, in order to optimize the space model for automated storage of finished cigarettes. The algorithm is run to obtain 20 Pareto solutions and examine their three objective functions. The experiment's findings revealed, after optimizing the NSGA‐II algorithm in this study, the average reduction rate of shipping efficiency is 32%, the average reduction rate of shelf stability is 54%, and the average reduction rate of product correlation is about 77%, indicating that the algorithm optimization is highly effective.

show abstract

NSGA-II With Simple Modification Works Well on a Wide Variety of Many-Objective Problems

Cited by 37 publications

References 21 publications

The dilemma between eliminating dominance-resistant solutions and preserving boundary solutions of extremely convex Pareto fronts

The dilemma between eliminating dominance-resistant solutions and preserving boundary solutions of extremely convex Pareto fronts

Solution representation learning in transfer evolutionary optimization

NSGA‐II algorithm‐based automated cigarette finished goods storage level optimization research

Contact Info

Product

Resources

About