Information-Modeling Technology for the Environmental Monitoring - New Ecoinformatics Tools in Environmental Science: Applications and Decision-making - page 38

Gauss fit these equations to voltage-clamp data.

This algorithm, known as the damped least squares method, provides a numerical
solution to the problem of minimizing a function, generally nonlinear, over a space
of parameters of the function.
Subsequent development of these approaches to the solution of tasks related to
biological evolution simulation is based on parameterization of basic evolution
factors

Newton algorithm, is often used to

-

hereditary mutability, competition, and selection. Progress in this direction
was achieved due to the introduction of structured model (C-model) into evolu-
tionary schemes (Bukatova and Makrusev 2003, 2004a, b).
Bukatova and Matveev (2004b) developed an idea of neuron networks for tasks
when the processing images are needed. In this case there are three possibilities for
neuron network selection:

—

1. Neuron network as finite-automat strati ed model or Petri net (Ren é and

Hassane 2005).
2. Modi

cation of NeuroShell 2, Neural Analyzer, and BrainMaker Pro depending
on the concrete neuro-paradigm (Lowd 1977).
3. Integrity-evolutionary integration medium that includes tools for structural

search of effective con guration of neuron net.

guration for
neuron network. Let input measurements x c (t) from finite-dimensional space X are
transformed to the solutions y c (t) from

Each of case includes the tools for structural search of ef cient con

finite-dimensional space Y by means of
neuron network that realizes unknown transformer F. In this case it is needed to
synthesis F: X → Y. This operation can be realized by means of evolutionary
synthesis algorithm represented in Figs. 1.15 and 1.16 . This algorithm uses neurons
and connections between them. Quantity criterion de

nes the compromise between
network complexity and precision results. This compromise is achieved in frame-
work of structural evolutionary synthesis that is based on stochastic search of C-
models. Variety of possible situations is considered by Bukatova and Makrusev
(2004a, b). Particularly, Krapivin et al. (2008a) considering the problem of infor-
mation-measuring system synthesize use a modern technologies and element base of
microelectronics and optoelectronics jointly with adaptive-evolutionary technology.
Gulyaev et al. (1987, 1989) developed technology of evolutionary computer that is
based on the optical electronics element base including discrete logical elements.

1.5 A Global Model as Unit of the Information-Modeling
Technology

1.5.1 Principal Structure of Global Model

Solution of global environmental problems is possible by means of the GIMS
technology that parameterizes processes in the nature/society system (NSS) of
global scale. There exist various approaches to the synthesis of a global model to