Self learning techniques

Apart from the actual measuring devices data processing is a key feature in this respect. These t) e of data are new. They are numerous with multiple complex interrelationships and time dependent. New self learning techniques will be required. The introduction of such an approach will be slow, because no evidence of the benefits are available at this moment. 

Hence zeolite synthesis shares some of the same combinatorial self organization and self learning aspects as seen in the antibody system. During zeolite crystallization template molecules are incorporated in unique positions. Molecular mechanics techniques have been developed ] that allow the prediction of optimum template zeohte interaction useful to select template molecules to synthesize preselected zeohte structmes. 

Now, one may ask, what if we are going to use Feed-Forward Neural Networks with the Back-Propagation learning rule Then, obviously, SVD can be used as a data transformation technique. PCA and SVD are often used as synonyms. Below we shall use PCA in the classical context and SVD in the case when it is applied to the data matrix before training any neural network, i.e., Kohonen s Self-Organizing Maps, or Counter-Propagation Neural Networks. 

In the present case, the electron hopping chemistry in the polymeric porphyrins is an especially rich topic because we can manipulate the axial coordination of the porphyrin, to learn how electron self exchange rates respond to axial coordination, and because we can compare the self exchange rates of the different redox couples of a given metallotetraphenylporphyrin polymer. To measure these chemical effects, and avoid potentially competing kinetic phenomena associated with mobilities of the electroneutrality-required counterions in the polymers, we chose a steady state measurement technique based on the sandwich electrode microstructure (19). 

---