Genetic Algorithms: A New Approach to the Timetable Problem

1992 ◽  
pp. 235-239 ◽  
Author(s):  
Alberto Colorni ◽  
Marco Dorigo ◽  
Vittorio Maniezzo
Keyword(s):  
Genetic Algorithms ◽  
New Approach ◽  
Timetable Problem

New approach to constrained shape optimization using genetic algorithms

AIAA Journal ◽  
1998 ◽  
Vol 36 ◽  
pp. 51-61 ◽  
Author(s):  
M. C. Sharatchandra ◽  
Mihir Sen ◽  
Mohamed Gad-el-Hak
Keyword(s):  
Genetic Algorithms ◽  
Shape Optimization ◽  
New Approach

scholarly journals When Darwin Met Einstein: Gravitational Lens Inversion with Genetic Algorithms

2005 ◽  
Vol 22 (2) ◽  
pp. 128-135 ◽  
Author(s):  
Brendon J. Brewer ◽  
Geraint F. Lewis
Keyword(s):  
Genetic Algorithms ◽  
Gravitational Lensing ◽  
Surface Brightness ◽  
Initial Study ◽  
Brightness Distribution ◽  
Gravitational Lens ◽  
Atmospheric Effects ◽  
Lens Model ◽  
New Approach ◽  
Genetic Algorithm Approach

AbstractGravitational lensing can magnify a distant source, revealing structural detail which is normally unresolvable. Recovering this detail through an inversion of the influence of gravitational lensing, however, requires optimisation of not only lens parameters, but also of the surface brightness distribution of the source. This paper outlines a new approach to this inversion, utilising genetic algorithms to reconstruct the source profile. In this initial study, the effects of image degradation due to instrumental and atmospheric effects are neglected and it is assumed that the lens model is accurately known, but the genetic algorithm approach can be incorporated into more general optimisation techniques, allowing the optimisation of both the parameters for a lensing model and the surface brightness of the source.

Optical Satellite Multiobjective Genetic Algorithms for Investigation of MH370 Debris

2020 ◽  
pp. 166-207
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Cloud Cover ◽  
Optical Remote Sensing ◽  
New Approach ◽  
Precise Method ◽  
Spectral Signatures ◽  
Remote Sensing Technique ◽  
Multiobjective Genetic Algorithms ◽  
Strength Constraints

At present, there is no precise method that can inform where the lost flight MH370 is. This chapter proposes a new approach to search for the missing flight MH370. To this end, multiobjective genetic algorithms are implemented. In this regard, a genetic algorithm is taken into consideration to optimize the MH370 debris that is notably based on the geometrical shapes and spectral signatures. Currently, there may be three limitations to optical remote sensing technique: (1) strength constraints of the spacecraft permit about two hours of scanning consistently within the day, (2) cloud cover prevents unique observations, and (3) moderate information from close to the ocean surface is sensed through the scanner. Needless to say that the objects that are spotted by different satellite data do not scientifically belong to the MH370 debris and could be just man-made without accurate identifications.

New Approach to Constrained Shape Optimization Using Genetic Algorithms

AIAA Journal ◽  
10.2514/2.351 ◽  
1998 ◽  
Vol 36 (1) ◽  
pp. 51-61 ◽  
Author(s):  
M. C. Sharatchandra ◽  
Mihir Sen ◽  
Mohamed Gad-El-Hak
Keyword(s):  
Genetic Algorithms ◽  
Shape Optimization ◽  
New Approach

FEATURE SELECTION FOR SUPPORT VECTOR MACHINES USING GENETIC ALGORITHMS

2004 ◽  
Vol 13 (04) ◽  
pp. 791-800 ◽  
Author(s):  
HOLGER FRÖHLICH ◽  
OLIVIER CHAPELLE ◽  
BERNHARD SCHÖLKOPF
Keyword(s):  
Genetic Algorithms ◽  
Feature Selection ◽  
Support Vector Machines ◽  
Cross Validation ◽  
Support Vector ◽  
Generalization Error ◽  
New Approach ◽  
Vector Machines ◽  
Selection For ◽  
Natural Way

The problem of feature selection is a difficult combinatorial task in Machine Learning and of high practical relevance, e.g. in bioinformatics. Genetic Algorithms (GAs) offer a natural way to solve this problem. In this paper we present a special Genetic Algorithm, which especially takes into account the existing bounds on the generalization error for Support Vector Machines (SVMs). This new approach is compared to the traditional method of performing cross-validation and to other existing algorithms for feature selection.

Sign in / Sign up

Export Citation Format

Share Document