Three Links Related to Self-Organizing Networks

Email.Email weblog link
Blog this.Blog this

Lucas Gonze
Jan. 22, 2002 03:40 PM

Atom feed for this author. RSS 1.0 feed for this author. RSS 2.0 feed for this author.

Capability-based Financial Instruments

"Smart Contracts: Patterns of Cooperation without Vulnerability

A contract is a mutually agreeable arrangement of rules among mutually suspicious parties so they may cooperate with limited risks to each other's mischief. It is a game both are willing to play because both expect to win."

Smart contracts are about the interfaces between pairs of nodes.


Self-Organizing Maps

"The basic Self-Organizing Map (SOM) can be visualized as a sheet-like neural-network array (see Figure 1), the cells (or nodes) of which become specifically tuned to various input signal patterns or classes of patterns in an orderly fashion. The learning process is competitive and unsupervised, meaning that no teacher is needed to define the correct output (or actually the cell into which the input is mapped) for an input. In the basic version, only one map node (winner) at a time is activated corresponding to each input. The locations of the responses in the array tend to become ordered in the learning process as if some meaningful nonlinear coordinate system for the different input features were being created over the network"

SOM is about how relationships between pairs of nodes affect the overall landscape.


Adaptive Landscapes

"[Sewell] Wright envisioned different stages of evolution [including] intrademic selection where selection within local populations (demes) would drive the various demes to the top of their nearest peak. Even if several populations were at different "locations" on the adaptive landscape, the highest peak may not be reached. One can invoke stage 2.5 by saying that drift in local populations might move such a population's allele frequency off one peak and into the 'domain of attraction' of an adjacent peak with a different maximum fitness. This peak may be lower or higher than the old one, but after several rounds of drift at least one population may evolve to the top of the highest peak."

Adaptive landscapes are about how topographies formed by algorithms like SOM are affected by the properties of things that algorithms like smart contracts control.

In the above graph, the vertical axis is about greater or lesser fitness. That is the same axis as in SOM diagrams. The other two axes are about physical characteristics that affect fitness; they're the kinds of things that smart contracts control. (Strictly speaking, smart contracts are about phenotype, the actual physical characteristics that happen as a result of genotypes and are the interface between our genes and the external world).

The part that is relevant to self-organizing networks is the way positioning on the phenotype axes implies both topography and form. If you are a node in a high bandwidth neighborhood then you get access to more bandwidth yourself, and you rise in bandwidth fitness. Nodes next to each other on the phenotype axes are both connected and similar.


Lucas Gonze works on Webjay, XSPF, and a survey of playlist formats.