Genetic algorithms essentials

The fourth example, with its six ways, leans even more toward provocadon while still being reasonably informative. The actual presentadon for which this served as a tide is about the use of six elements of practical GA theory to make genetic algorithms work better in applicadons, and this dde does hint at that, creating interest by shrouding the six ways in mystery. (If you re interested in drawing an audience to a presentation of theory, some mystery in one s title is essential.)... 

The program is reported to carry out simple Hiickel molecular orbital calculations to determine the relative sensitivity of aromatic carbon atoms to oxidation and the relative stability of keto and enol tautomers. Klopman et al. (1999) have reported that for polycyclic aromatic hydrocarbons, adequate reactivity is an essential but not sufficient condition for enzyme catalyzed reaction. The accessibility of the reactive site (i.e., the absence of steric hindrance) was also found to be important. Genetic algorithms have been used to optimize the performance of the biotransformation dictionary by treating the initial priority scores set by expert assessment as adjustable parameters (Klopman et al., 1997). 

Stochastic search methods offer a robust quality to optimisation processes. The most widely used stochastic search methods in the literature include genetic algorithms (GA), evolutionary strategies (ES), simulated annealing (SA) and tabu search (TS). The GA and ES are essentially the same (initially the former focused on discrete variables and the latter focused on continuous variables). They emulate nature s evolutionary behaviour, and the search evolves throughout... 

Genetic algorithm (GA) is a kind of widely used method to determine the global minimum structure of a cluster. It can sample the potential energy surface efficiently and hop from one region of the PES to another region rather easily. It is inspired by Darwinian evolution theory that only the fittest individuals can survive. The basic philosophy of GA is to mimic the natural selection and evolution processes in nature. The essential idea of GA procedure is to allow a population of a number of individual candidates to evolve under a given selection rule that maximizes the fitness function. 

---