Competitive learning model

Training a Kohonen neural network with a molecular descriptor and a spectrum vector models the rather complex relationship between a molecule and an infrared spectrum. This relationship is stored in the Kohonen network by assigning the weights through a competitive learning technique from a suitable training set of... 

A subset of VQ algorithms comprises Competitive Learning (CL) methods, where a neural network model is used to find an approach of VQ calculation in an unsupervised way. Their advantage over other VQ algorithms is that CL is simple and easily parallelizable. Well known CL approaches are K-means [13] (including... 

In order to demonstrate that uncomplexed bromine atoms act as chain propagators, toluene and ethylbenzene were photobromina-ted in a competition study at pressures of 75 to 423 bar and at 40 °C. Over the entire pressure range, the reactivity of the benzylic secondary C-H bond in ethylbenzene was found to be about 30 times greater than that of the corresponding primary C-H bond in toluene. The analogous value for the reactivity in CCI4 at 40 °C is 36. The bromine atoms in SC-CO2 are therefore particularly free. It would be important to determine quantum yields (chain lengths) at various pressures to learn more about mechanistic aspects and other details of the reaction. Local solvent structures on model free-radical reactions in SC-CO2 have been analyzed in some detail. [33]... 

---