We propose a multi-strategy genetic algorithm for performing belief revision. The algorithm implements a new evolutionary strategy which is a combination of the theories of Darwin and Lamarck. Therefore, the algorithm not only includes the Darwinian operators of selection, mutation and crossover but also a Lamarckian operator that changes the individuals so that they perform better in solving the given problem. This is achieved through belief revision directed mutations, oriented by tracing logical derivations. The algorithm, with and without the Lamarckian operator, is tested on a number of belief revision problems, and the results show that the addition of the Lamarckian operator improves the eciency of the algorithm. We believe that the combination of Darwinian and Lamarckian operators will be useful not only for standard belief revision problems but especially for problems where the chromosomes may be exposed to dierent constraints and observations. In these cases, the Lamarckian and Darwinian operators would play a dierent role: the Lamarckian one would be used in order to bring a chromosome closer to a solution or to nd an exact solution of the current belief revision problem, while Darwinian ones will have the aim of preparing chromosomes to deal with new situations by exchanging genes among them.

Logic aided Lamarckian evolution

LAMMA, Evelina;RIGUZZI, Fabrizio
2000

Abstract

We propose a multi-strategy genetic algorithm for performing belief revision. The algorithm implements a new evolutionary strategy which is a combination of the theories of Darwin and Lamarck. Therefore, the algorithm not only includes the Darwinian operators of selection, mutation and crossover but also a Lamarckian operator that changes the individuals so that they perform better in solving the given problem. This is achieved through belief revision directed mutations, oriented by tracing logical derivations. The algorithm, with and without the Lamarckian operator, is tested on a number of belief revision problems, and the results show that the addition of the Lamarckian operator improves the eciency of the algorithm. We believe that the combination of Darwinian and Lamarckian operators will be useful not only for standard belief revision problems but especially for problems where the chromosomes may be exposed to dierent constraints and observations. In these cases, the Lamarckian and Darwinian operators would play a dierent role: the Lamarckian one would be used in order to bring a chromosome closer to a solution or to nd an exact solution of the current belief revision problem, while Darwinian ones will have the aim of preparing chromosomes to deal with new situations by exchanging genes among them.
2000
Logic Programming; Genetic Algorithms; Belief Revision
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11392/1195323
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact