DCS Paper

This paper describes some biologically-oriented processes that could be used to build the sort of networks we associate with the human brain. A ‘refined’ neuron will be proposed that can process more variable sets of values, but with the same level of control and reliability that a binary neuron would have. When modelling the human brain on a computer, it would be normal to consider using binary units, but there is a sense that our intelligence should derive from something more complex than this. There is nothing particularly new in this paper with regard to basic structures, but after reading, it will be possible to think of an essentially binary system in terms of a more variable set of values that can be created and reliably controlled. A plausible automatic construction of such a network will be proposed. This will allow for more complex formulae or equations to be considered as normal. The paper also shows how most of the recent research results can be part of a more complete model. These are largely in line with conventional thinking, but possibly with one or two more radical suggestions. The intention is to describe plausible mechanical processes that have some mathematical foundation, but that could be used without the magical ‘intelligence’ part that is still not fully understood. There are some important updates from an earlier version of this paper.