D. W. Corne, J. D. Knowles, and M. J. Oates. The Pareto Envelope-based Selection Algorithm for Multiobjective Optimization. In Proceedings of the Parallel Problem Solving from Nature VI Conference, pages 839--848. Springer, 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....in the recent past, mainly because of their ability to find a wide spread of Paxeto optimal solutions in a single simulation run. Vaxious evolutionary approaches to multi objective optimization have been proposed since 1985. Some of fairly recent ones axe NSGA II [9] SPEA2 [19] PESA [1] (which axe included in this study) and others. They all have been mainly applied to two to three objectives. In order to establish their superiority over classical methods and demonstrate their abilities for convergence and maintenance of diversity, they need to be tested on higher number of ....

....function (governed by the reduction rate r) to introduce more diversity into the population. NSGA II, as reported in [8] outperformed PAES in preserving the spread of non dominated front on five 2 objective test problems (listed in [8] PESA (Pareto Enveloped based Selection Algorithm) [1], an improvement of PAES, uses the hyper cubes grid division not only for crowding as in PAES, but also for selection process. PESA was compared with PAES and SPEA on six test functions T to To [2] each of which is a 2 objective problem defined on m parameters) and was reported to outperform SPEA ....

[Article contains additional citation context not shown here]

D. W. Come, J. D. Knowles, and M. J. Oates. The Pareto Envelope-based Selection Algorithm for Multiobjective Optimization. In M. Schoenauer, K. Deb, G. Rudolph, X. Yao, E. Lutton, J. J. Merelo, and H.-P. Schwefel, editors, Proceedings of the Parallel Problem Solving from Nature VI Conference, pages 839-848, Paxis, France, 2000. Springer. Lecture Notes in Computer Science No. 1917.

....then, many Evolutionary algorithms for solving multi objective optimization problems have been developed. The most recent ones are the Non Dominated Sorting Genetic Algorithm II (NSGA II) 3] Strength Pareto Evolutionary Algorithm II (SPEA II) 16] Pareto Envelope based selection II (PESA II) [17]. Most of these approaches propose the use of a generational GA. Deb proposed an Elitist Steady State Multi objective Evolutionary Algorithm (MOEA) 18] which attempts to maintain spread [15] while attempting to converge to the true Pareto optimal front. This algorithm requires sorting of the ....

Crone, D. W., Knowles, J.D., and Oates, M.J. (2000). The Pareto Envelope-based Selection Algorithm for Multi-objective Optimization. In Schoenauer, M., Deb, K., Rudolph, g., Yao, X., Luton, E., Merelo, J.J., and Schewfel, H.-P., editors, Proceedings of the Parallel Problem Solving from Nature VI Conference, pp.839-848, Paris, France. Springer. Lecture Notes in Computer Science No. 1917.

....for a controlled elitism in evolutionary multi objective optimization, demonstrated in this paper should encourage similar or other ways of implementing controlled elitism in other multi objective evolutionary algorithms. 1 Introduction It is now well established through a number of studies [1, 11] that elitist multiobjective evolutionary algorithms (MOEAs) have better convergence characteristics than non elitist MOEAs. Motivated by these studies, researchers and practitioners now concentrate on developing and using elitist MOEAs. This have resulted in a number of elitist MOEAs, such as ....

Corne, D. W., Knowles, J. D., and Oates, M. J. (2000). The Pareto envelopebased selection algorithm for multiobjective optimization. Proceedings of the Parallel Problem Solving from Nature VI Conference, pp. 839-848.

....optimization problems. In different studies (Zitzler and Thiele 1999; Zitzler, Deb, and Thiele 2000) SPEA has shown very good performance in comparison to other multiobjective evolutionary algorithms, and therefore it has been a point of reference in various recent investigations, e.g. (Corne, Knowles, and Oates 2000). Furthermore, it has been used in different applications, e.g. Lahanas, Milickovic, Baltas, and Zamboglou 2001) In this paper, an improved version, namely SPEA2, is proposed, which incorporates in contrast to its predecessor a fine grained fitness assignment strategy, a density estimation ....

.... that were extensively compared to several existing evolution based methods (Zitzler and Thiele 1999; Zitzler, Deb, and Thiele 2000) As it clearly outperformed the (nonelitist) alternative approaches under consideration, it has been used as a point of reference by various researchers, e.g. (Corne, Knowles, and Oates 2000; Jaszkiewicz 2000; Tan, Lee, and Khor 2001) Meanwhile further progress has been made and recently proposed methods, for instance NSGA II (Deb, Agrawal, Pratap, and Meyarivan 2000) and PESA (Corne, Knowles, and Oates 2000) were shown to outperform SPEA on certain test problems. Furthermore, new ....

[Article contains additional citation context not shown here]

Corne, D. W., J. D. Knowles, and M. J. Oates (2000). The pareto envelope-based selection algorithm for multiobjective optimisation. In M. S. et al. (Ed.), Parallel Problem Solving from Nature -- PPSN VI, Berlin, pp. 839--848. Springer.

....selection is performed on the union of population and external set and recombination and mutation operators are applied as usual. As SPEA has shown very good performance in different comparative studies [ZT99, ZDT00] it has been a point of reference in various recent investigations, e.g. [CKO00]. Furthermore, it has been used in different applications, e.g. LMBoZ01] 2 1 4 3 2 3 identical fitness F F 1 identical fitness F F identical fitness F F identical fitness F f1 f2 Figure 1: Illustration of SPEA s fitness assignment scheme in the case of a highly discretized ....

D. W. Corne, J. D. Knowles, and M. J. Oates. The pareto envelope-based selection algorithm for multiobjective optimisation. In Marc Schoenauer et al., editor, PPSN VI, pages 839--

....decision variables) and by updating the subspaces dynamically. Authors have later proposed an improved method such as Pareto envelope based selection algorithm or PESA, which uses the subpopulation stored in each grid location to control the selection pressure and diversity of population members [CKO00] The above algorithm clearly prefers non dominated solutions, an essential property for an MOEA to progress towards the true Pareto optimal front. The diversity is also maintained by the deterministic crowding approach. However, the algorithm lacks a convergence proof, simply because a ....

D. Corne, J. Knowles, and M. Oates. The Pareto envelope-based selection algorithm for multiobjective optimization. In Proceedings of the Sixth International Conference on Parallel Problem Solving from Nature VI (PPSN-VI), pages 839--848, 2000.

....intuitive: Though the archive must be truncated, one would like it to be as diverse as possible. Hence, individuals in a densely populated area receive lower values and are discarded from the archive in favor of others. Different implementations of this concept are applied in [12] 7] 8] [2], or [3] In this study we represent the method used in NSGA II [3] by the operator truncate # : For each objective coordinate the absolute difference of its predecessor and successor is aggregated for each individual, higher total values lead to better ranks. Table 2 gives an overview of these ....

....baseline, the truncate # operator is included, which represents the unlimited archive. Method No Reduction Conservative Random Clustering Density based Operator truncate# truncate# truncate# truncate# truncate# Examples VV [13] AR 1 [13] SPEA [20] PAES [ 7 ] PR [13] AR 2 [13] M PAES [8] PESA [2] NSGA II [3] Features may grow efficiency easy good good very large, preserving, implementation, discrimination, discrimination, genetic drift unreachable low complexity, adaptive metrics, medium points genetic drift high complexity complexity Tab l e 2 . Archive truncation methods in ....

D. W. Corne, J. D. Knowles, and M. J. Oates. The pareto envelope-based selection algorithm for multiobjective optimisation. In Marc Schoenauer et al., editor, Parallel Problem Solving from Nature -- PPSN VI, Berlin. Springer.

....elitism has much better convergence property than the original NSGA II. The suggested controlled elitism preserving approach is generic and can also be implemented with other elitist multi objective evolutionary algorithms. 1 Introduction It is now well established through a number of studies [11, 1] that elitist multiobjective evolutionary algorithms (MOEAs) have better convergence characteristics than non elitist MOEAs. Motivated by these studies, researchers and practitioners now concentrate on developing and using elitist MOEAs. This have resulted in a number of elitist MOEAs, such as ....

Corne, D. W., Knowles, J. D., and Oates, M. J. (2000). The Pareto envelopebased selection algorithm for multiobjective optimization. Proceedings of the Parallel Problem Solving from Nature VI Conference, pp. 839-848.

No context found.

D. W. Corne, J. D. Knowles, and M. J. Oates. The Pareto Envelope-based Selection Algorithm for Multiobjective Optimization. In Proceedings of the Parallel Problem Solving from Nature VI Conference, pages 839--848. Springer, 2000.

No context found.

David W. Corne and Joshua D. Knowles. The Pareto-envelope based selection algorithm for multiobjective optimization. In Proceedings of the Sixth International Conference on Parallel Problem Solving from Nature (PPSN VI), pages 839-848, Berlin, 2000. Springer-Verlag.

....i 1 return (N) Termination Figure 2: AESSEA pressure of the algorithm is not too strong since selection for mating is purely random, and offspring only replace one of their parents, rather than the weakest member of the population. Some testing of this algorithm and comparison with PESA [3] suggest that it is both an effective and computationally efficient, multiobjective EA. In AESSEA, the function rand( returns a uniformly distributed deviate in [0; 1) and the function rand mem(P ) returns with uniform probability a member of the current population, P . The function archive(c) ....

D. W. Corne and J. D. Knowles. The Paretoenvelope based selection algorithm for multiobjective optimization. In Proceedings of the Sixth International Conference on Parallel Problem Solving from Nature (PPSN VI), pages 839--848, Berlin, 2000. Springer-Verlag.

....and offspring compete, but the overall selection pressure of the algorithm is not too strong since selection for mating is purely random, and offspring only replace one of their parents, rather than the weakest member of the population. Some testing of this algorithm and comparison with PESA [2] suggest that it is both an effective and computationally efficient, multiobjective EA. RPM decoder encoding The randomized primal method was put forward in [6] where it is described in detail) as an encoding for solving the d MST problem using any metaheuristic search method. It is a decoder ....

D. W. Corne and J. D. Knowles. The Pareto-envelope based selection algorithm for multiobjective optimization. In Proceedings of the Sixth International Conference on Parallel Problem Solving from Nature (PPSN VI), pages 839--848, Berlin, 2000. Springer-Verlag.

....solution if it is not worse than any solution found so far. In the context of PAES, however, worse means dominated. The second pair of algorithms, which are a mutation only genetic algorithm with deterministic crowding [11] DCGA) and the Pareto envelope based selection algorithm (PESA) [2], are both multi point hill climbers (neither uses recombination here) Once again, they are supposed to be analogues of each other, subject to the differences forced upon them by the different requirements of single and multiple objective optimization. The analogy between them is not as clear as ....

....one from each parent. Parents and offspring are then paired up so as to minimize the sum of the genotypic Hamming distance between them. Each offspring then replaces the parent it is paired with if it non worse than that parent. The PESA algorithm used here has been described in detail in [2], and pseudocode for it is given in Figure 3. It has an internal population IP of size P I , and an external population of nondominated solutions EP . Here it is used without crossover so that each generation consists of selecting P I parents from EP and mutating them to produce P I new offspring. ....

D. W. Corne and J. D. Knowles. The Pareto-envelope based selection algorithm for multiobjective optimization. In Proceedings of the Sixth International Conference on Parallel Problem Solving from Nature (PPSN VI), pages 839--848, Berlin, 2000. Springer-Verlag.

No context found.

Corne, D.W., Knowles, J.D., and Oates, M.J. (2000). The Pareto envelope-based selection algorithm for multiobjective optimization. Proceedings of the Parallel Problem Solving from Nature VI Conference, pp. 839-848.

No context found.

D. W. Corne, J. D. Knowles, and M. J. Oates. The pareto envelope-based selection algorithm for multiobjective optimisation. In Marc Schoenauer et al., editor, Parallel Problem Solving from Nature - PPSN VI, pages 839-848, Berlin. Springer, (2000).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC