Re: Terence Tao • Modular Arithmetic Challenge

• Can a neural network learn to do modular multiplication efficiently?

Incidental Reflection 1

There are alternative models of neural networks which do not depend on threshold neurons and endlessly fiddling with weights.

Incidental Reflection 2

The following three blog posts present a case study comparing two ways of handling a classic example from the Parallel Distributed Processing paradigm, the “Jets and Sharks” database problem, first, the original treatment of McClelland and Rumelhart and, second, according to a program I developed all through the 1980s.  The latter method uses ideas from Grossberg’s competition‑cooperation and winner‑take‑all dynamics, but is purely syntactic and propositional logic based, involving no extraneous weights.

• Theme One Program • Jets and Sharks • (1) • (2) • (3)

Resource

• Survey of Theme One Program

cc: Academia.edu • Cybernetics • Laws of Form • Ma^thstodon
cc: Research Gate • Structural Modeling • Systems Science • Syscoi