Learning growing cell structure

The growing cell structure algorithm is a variant of a Kohonen network, so the GCS displays several similarities with the SOM. The most distinctive feature of the GCS is that the topology is self-adaptive, adjusting as the algorithm learns about classes in the data. So, unlike the SOM, in which the layout of nodes is regular and predefined, the GCS is not constrained in advance to a particular size of network or a certain lattice geometry. 

Fritzke, B. (1994) Growing cell structures-a self-organizing network for unsupervised and supervised learning. Neural Networks 7 1441-1460... 

We have examined how cells maintain close control of CDK activity, but how does the activity of CDK control the cell cycle The list of target proteins that CDKs are known to act upon continues to grow, and much remains to be learned. But we can see a general pattern behind CDK regulation by inspecting the effect of CDKs on the structures of laminin and myosin and on the activity of retinoblastoma protein. 

The structure of a module about the hydrogen car is briefly presented in Fig. 1. The content knowledge deals with redox reactions and the design and function of fuel cells. The students thereby learn about a technique that is already more than 100 years old but has only recently become promising again for societal reasons the growing demand for mobility cannot be fulfilled anymore because of the shortage of crude oil and fossil fuels in the future. 

The structure of this book aims to emphasize the similarity of fuel cells of different types. Nowadays, when the number of publications on FC science and technology is growing at an alarming rate, the fuel cell community tends to separate into sub-domains interested in only one type of cell. In an attempt to counteract this unfortunate trend the book is structured according to the hierarchical principle (from catalyst layers to stacks), rather than according to the cell type. SOFC, PEFC and DMFC communities can learn a lot from each other. 

---