Networks growing according to the rule that every new node has a probability p k of being attached to k preexisting nodes, have a universal phase diagram and exhibit power-law decays of the distribution of cluster sizes in the non-percolating phase. The percolation transition is continuous but of in
Synthesis of universal networks
β Scribed by A. M. Romankevich
- Publisher
- Springer US
- Year
- 1971
- Tongue
- English
- Weight
- 177 KB
- Volume
- 7
- Category
- Article
- ISSN
- 1573-8337
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
## This paper presents two active RC networks-one containing finite-gain amplifiers and the other being the differential amplifier version-each of which can be used to realize the ratio of arbitrary polynomials in s as any of the following five two-port parameters: Voltage transfer function, drivi
Hava Siegelmann and Eduardo Sontag have shown that recurrent neural networks using the linear-bounded sigmoid are computationally universal. We show that this remains true if the linear-bounded sigmoid is replaced by any function in a fairly large class.
Given the class of symmetric discrete weight neural networks with finite state set (0, l}, we prove that there exist iteration modes under these networks which allow to simulate in linear space arbitrary neural networks (non-necessarily symmetric). As a particular result we prove that an arbitrary s
Fluid neural networks can be used as a theoretical framework for a wide range of complex systems as social insects. In this article we show that collective logical gates can be built in such a way that complex computation can be possible by means of the interplay between local interactions and the c