Category: GA
Chapter: 4
Comment: ヒューリスティックを使って初期値をきめるとバリエーショ
        ンが小さくなるのでまずい。クロスオーバーやミューテー
        ションはアプリケーション毎に工夫するのがよい。最終的
        に良い結果を得るにはGAで得られた結果にローカル最適化
        手法を適用するのがよい。
Abstract: Like other weak methods, Genetic Algorithms (GA's)
        are applicable to a broad range of problems for
        which very little prior knowledge is available.
        However, many opportunities exist for incorporationg
        available problem specific heuristics into GA-based
        systems. This paper explores some of these
        opportunities in the context of the traveling
        salesperson problem. In particular, several
        heuristic methods for population initialization,
        crosover and mutation are discussed and empirical
        comparisons are presented.
Bibtype: InBook
Booktitle: Genetic Algorithms and Simulated Annealing
Author: John J. Grefenstette
Pages: 42-60
Title: Incorporating Problem Specific Knowledge into
        Genetic Algorithm
Editor: Lawrence Davis
Year: 1987
Publisher: Morgan Kaufmann Publishers