Application of artificial intelligence to search ground-state geometry of clusters

We introduce a global optimization procedure, the neural-assisted genetic algorithm (NAGA). It combines the power of an artificial neural network (ANN) with the versatility of the genetic algorithm. This method is suitable to solve optimization problems that depend on some kind of heuristics to limit the search space. If a reasonable amount of data is available, the ANN can "understand" the problem and provide the genetic algorithm with a selected population of elements that will speed up the search for the optimum solution. We tested the method in a search for the ground-state geometry of silicon clusters. We trained the ANN with information about the geometry and energetics of small silicon clusters. Next, the ANN learned how to restrict the configurational space for larger silicon clusters. For Si 1 0 and Si 2 0 , we noticed that the NAGA is at least three times faster than the "pure" genetic algorithm. As the size of the cluster increases, it is expected that the gain in terms of time will increase as well.

[1]  Ho,et al.  Molecular geometry optimization with a genetic algorithm. , 1995, Physical review letters.

[2]  P. Cao,et al.  Stable structures for Si 20 clusters , 2000 .

[3]  A. Shvartsburg,et al.  Ionization of medium-sized silicon clusters and the geometries of the cations , 1998 .

[4]  Thomas R. Cundari,et al.  A Comparison of Neural Networks versus Quantum Mechanics for Inorganic Systems , 1997, J. Chem. Inf. Comput. Sci..

[5]  Bernd Hartke,et al.  Global cluster geometry optimization by a phenotype algorithm with Niches: Location of elusive minima, and low‐order scaling with cluster size , 1999 .

[6]  L. Mitas,et al.  Family of low-energy elongated Sin (n <= 50) clusters. , 1995, Physical review. B, Condensed matter.

[7]  Bicai Pan,et al.  Structures of medium-sized silicon clusters , 1998, Nature.

[8]  M. R. Lemes,et al.  Generalized simulated annealing: Application to silicon clusters , 1997 .

[9]  M. R. Lemes,et al.  Combining genetic algorithm and simulated annealing: A molecular geometry optimization study , 1998 .

[10]  L. Mitas,et al.  Quantum Monte Carlo determination of electronic and structural properties of Sin clusters (n <= 20). , 1995, Physical review letters.

[11]  C. Miller,et al.  1990: annus mirabilis of potassium channels , 1991, Science.

[12]  L. Wille,et al.  Computational complexity of the ground-state determination of atomic clusters , 1985 .

[13]  Martin F. Jarrold,et al.  Nanosurface Chemistry on Size-Selected Silicon Clusters , 1991, Science.

[14]  J. Hopfield,et al.  Computing with neural circuits: a model. , 1986, Science.

[15]  Kolomenskii,et al.  Nonlinear excitation of capillary waves by the Marangoni motion induced with a modulated laser beam. , 1995, Physical review. B, Condensed matter.

[16]  K. Ho,et al.  Structural optimization of Lennard-Jones clusters by a genetic algorithm , 1996 .

[17]  Kaxiras,et al.  Shape of small silicon clusters. , 1993, Physical review letters.

[18]  Structural models for intermediate-sized Si clusters , 1993 .