PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Divide-and-Evolve: a New Memetic Scheme for Domain-Independent Temporal Planning
Marc Schoenauer, Pierre Savéant and VIncent Vidal
In: EvoCOP 2006, 10-12 Apr 2006, Budapest (Hungaria).

Abstract

An original approach, termed Divide-and-Evolve is proposed to hybridize Evolutionary Algorithms (EAs) with Op erational Research (OR) methods in the domain of Temp oral Planning Problems (TPPs). Whereas standard Memetic Algorithms use local search methods to improve the evolutionary solutions, and thus fail when the local method stops working on the complete problem, the Divide-and-Evolve approach splits the problem at hand into several, hop efully easier, sub-problems, and can thus solve globally problems that are intractable when directly fed into deterministic OR algorithms. But the most prominent advantage of the Divide-and-Evolve approach is that it immediately op ens up an avenue for multi-ob jective optimization, even though the OR method that is used is single-ob jective. Proof of concept approach on the standard (single-ob jective) Zeno transportation benchmark is given, and a small original multi-ob jective benchmark is prop osed in the same Zeno framework to assess the multi-objective capabilities of the prop osed methodology, a breakthrough in Temporal Planning.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
ID Code:2732
Deposited By:Marc Schoenauer
Deposited On:22 November 2006