Orvosh, D. and L. Davis (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In S. Forrest (Ed.), Proceedings of the 5 International Conference on Genetic Algorithms, pp. 650. Morgan Kaufmann.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....not generalize. Other techniques that can be used to handle constraints are more or less problem dependent. For instance, the knowledge about linear constraints can be incorporated into specific operators [4] or a repair operator can be designed that projects infeasible points onto feasible ones [6]. Moreover, the experimental results reported in many papers suggest that making an appropriate a priori choice of an evolutionary method for a nonlinear parameter optimization problem remains an open question. It seems that the most promising approach at this stage of research is experimental, ....

Orvosh, D. and L. Davis (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In S. Forrest (Ed.), Proceedings of the 5 International Conference on Genetic Algorithms, pp. 650. Morgan Kaufmann.

....cial intelligence. On the theoretical side, Hinton and Nowlan (1987) addressed the issue of what information should be used from the local searcher. They analyzed the performance of two mechanisms namely, Baldwinian and Lamarckian learning. Their view of using the tness alone was supported by Orvosh and Davis (1993). Lobo and Goldberg (1997) addressed the issue of how to get the best of both searchers in the hybrid. Goldberg and Voessner (1999) used a simpli ed model of what they called an idealized hybrid algorithm, to analyze time partitioning, eciency and reliability of hybrid GAs. Local searchers are ....

....approach. One in which the local search is used during the tness evaluation and only the tness of the termination point under local search is used. This is a 100 Baldwinian approach. In practice, we use both Lamarckian and Baldwinian learning with a low percentage of Lamarckian steps. Orvosh and Davis (1993), has empirical results that suggest how often a Lamarckian step should be employed. 3 Sharing in Ordinary GAs vs. Hybrid GAs Many practical search and optimization problems require the investigation of multiple optima (Goldberg Richardson, 1987) The biological concept of niche formation in ....

Orvosh, D., & Davis, L. (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In Forrest, S. (Ed.), Proceedings of the Fifth International Conference on Genetic Algorithms (pp. 650). San Mateo, CA: Morgan Kaufmann.

....hybrids, but an important distinction between Baldwinian and Lamarckian learning was made by Hinton and Nowlan (1987) This issue of substituting the individual from the termination point of the local searcher into the population for further genetic search has been a much debated one. The study by Orvosh and Davis (1993) has interesting empirical results. A majority of the work has been focused on narrow application domains (combinatorial optimization, Traveling Salesman Problem, etc. A generic theoretical framework for combining GAs with other methods has been lacking. The study by Goldberg and Voessner ....

Orvosh, D., & Davis, L. (1993). Shall we repair? genetic algorithms, combinatorial optimization, and feasibility constraints. Proceedings of the Fifth International Conference on Genetic Algorithms, 650.

....generalize. Other techniques that can be used to handle constraints in are more or less problem dependent. For instance, the knowledge about linear constraints can be incorporated into speci c operators [24] or a repair operator can be designed that projects infeasible points onto feasible ones [30]. Section 1.2 provides a general overview of constraints handling methods for evolutionary computation techniques. The experimental results reported in many papers suggest that making an appropriate a priori choice of an evolutionary method for a nonlinear parameter optimization problem remains ....

Orvosh, D. and L. Davis (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In S. Forrest (Ed.), Proceedings of the 5 th International Conference on Genetic Algorithms, pp. 650. Morgan Kaufmann.

....easy to repair an unfeasible individual. Such repaired version can be used either for evaluation only, i.e. eval u (X) eval f (Y ) where Y is a repaired (i.e. feasible) version of X, or it can also replace the original individual in the population (with some probability) Recently (see Orvosh and Davis 1993) a so called 5 rule was reported: this heuristic rule states that in many combinatorial optimization problems, an evolutionary computation technique with a repair algorithm provides the best results when 5 of repaired individuals replace their unfeasible originals. However, the author is not ....

Orvosh, D. and L. Davis (1993). Shall We Repair? Genetic Algorithms, Combinatorial Optimization, and Feasibility Constraints. In Proceedings of the Fifth International Conference on Genetic Algorithms, Los Altos, CA, Morgan Kaufmann Publishers, 650.

....evolutionary algorithms provide an extremely convenient context in which to embed other techniques, and numerous approaches have been considered. The addition of specialised move operators incorporating heuristics or exploiting domain knowledge in other ways is often effective (Michalewicz, 1992; Davis Orvosh, 1993). If a true local optimisation algorithm exists, these ideas can be extended by applying full local optimisation to each child solution produced before evaluation, yielding a memetic algorithm (see chapter 9) Embedding a search technique in the genotype phenotype mapping is another possibility ....

....1992) exhibiting great sensitivity to the values of their many free parameters, and feeding rather too little information back to the algorithm to allow it to handle the constraints satisfactorily. While other methods are available for problems with explicit constraints (including repair methods, Davis Orvosh, 1993; smart decoders, Davis, 1987, 1991b; and special operators incorporating problem knowledge, Michalewicz Janikow, 1991) these do not have fully general applicability, and tend to require significant work for each new class of problems tackled. There is thus a need for a method that combines the ....

[Article contains additional citation context not shown here]

L. Davis and D. Orvosh, 1993. Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In S. Forrest, editor, Proceedings of the Fifth International Conference on Genetic Algorithms, page 50. Morgan Kaufmann (San Mateo).

....made between Baldwinian and Lamarckian learning by Hinton and Nowlan (1987) is particularly important and suggests that a local search method appended to a GEA can have a useful effect without backsubstituting the genotype corresponding to the termination point of the local search algorithm. Orvosh and Davis (1993) support this point of view with their empirically derived rule of 20 that suggests that a Lamarckian step if used at all should only be used one in twenty trials in order that population diversity not be overly disturbed. Another theoretical thread picked up in the literature is that of ....

Orvosh, D, & Davis, L. (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. Proceedings of the Fifth International Conference on Genetic Algorithms, 650.

....made between Baldwinian and Lamarckian learning by Hinton and Nowlan (1987) is particularly important and suggests that a local search method appended to a GEA can have a useful effect without backsubstituting the genotype corresponding to the termination point of the local search algorithm. Orvosh and Davis (1993) support this point of view with their empirically derived rule of 10 that suggests that a Lamarckian step if used at all should only be used one in ten trials in order that population diversity not be overly disturbed. Another theoretical thread picked up in the literature is that of adaptive ....

Orvosh, D, & Davis, L. (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. Proceedings of the Fifth International Conference on Genetic Algorithms, 650.

....P s (replacement process) where they undergo additional transformation by specialized operators. One of the most interesting parameters of the developed system is the probability of replacement p r (replacement of s by z in population of search points P s ; see figure 7) Recently, Orvosh and Davis (1993) reported a so called 5 rule: this heuristic rule states that in many combinatorial optimization problems, an evolutionary computation technique with a repair algorithm provides the best results when 5 of repaired individuals replace their infeasible originals. However, neither Davis (1995) nor ....

Orvosh, D. and L. Davis (1993). Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In S. Forrest (Ed.), Proceedings of the 5 th International Conference on Genetic Algorithms, pp. 650. Morgan Kaufmann.

....fitness is changed and the genotype remains unchanged. The drawback of the Lamarckian approach is lost of diversity. On the contrary, the Baldwinian strategy maintains the diversity in the population and can be very useful in a search space that doesn t have nice hills (Hinton Nowlan, 1987) Orvosh and Davis (1993) did an interesting combination of these two strategies for the Survivable Network Design Problem and verified that 95 of Baldwinian steps and 5 of Lamarckian steps performed best for their particular problem. This seems to indicate that a high proportion of Baldwinian steps is beneficial, but ....

Orvosh, D., & Davis, L. (1993). Shall we repair? Genetic algorithms, combinatorial optimizat ion, and feasibility constraints. Proceedings of the Fifth International Conference on Genetic Algorithms, 650.

....do alter in quite complex ways before they are mature enough to reproduce. used in the next generation. This is similar to the debate as to whether, in the context of constrained problems, to use an original but infeasible chromosome or the repaired feasible one. Recently, Orvosh and Davis [17] reports some interesting experimental evidence which suggests that this decision should itself be a stochastic choice. In this paper, rather than keeping the local optimization separate from recombination, we intend to explore the possibilities of integrating it directly into the standard genetic ....

D.Orvosh and L.Davis (1993) Shall we repair? Genetic algorithms, combinatorial optimization and feasibility constraints. In [37].

....the overall objective function; ffl Use problem specific information to stay within the feasible region. Michalewicz (1995) gives a detailed coverage of some popular techniques used for cBBO in the GA framework. Work in related areas can be found elsewhere (Homaifar, Lai, Qi, 1994; Jones, 1994; Orvosh Davis, 1993; Paredis, 1994; Powell Skolnick, 1993; Richardson, Palmer, Liepins, Hilliard, 1989; Schoenauer Xanthakis, 1993) Due to the lack of the theoretical footing, the basis of selecting one of the above methods over any other method is purely heuristic and experience. cSEARCH aims at developing a ....

Orvosh, D., & Davis, L. (1993). Shall we repair? genetic algorithms, combinatorial optimization, and feasibility constraints. See Forrest (1993), pp. 650.

No context found.

Lawrence Davis and David Orvosh, 1993. Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In Stephanie Forrest, editor, Proceedings of the Fifth International Conference on Genetic Algorithms. Morgan Kaufmann (San Mateo).

No context found.

Lawrence Davis and David Orvosh, 1993. Shall we repair? Genetic algorithms, combinatorial optimization, and feasibility constraints. In Stephanie Forrest, editor, Proceedings of the Fifth International Conference on Genetic Algorithms. Morgan Kaufmann (San Mateo).

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