PCS_convergence_McCandR {PCSinR}

Check a PCS network for convergence

Description

This function applies the convergence criterion defined by McClelland and Rumelhart to a given network, and returns either a (qualitative) boolean value that represents the convergence state, or a (quantitative) value that represents the number of iterations (of the last 10) that have met the convergence threshold.

Usage

PCS_convergence_McCandR(iteration, current_energy, memory.matrix,
  stability_criterion = 10^-6, output = "qualitative")

Arguments

iteration

The iteration to consider – in most cases, this will be the current iteration during a simulation run, however, the check can also be applied to a model output retroactively, and the iteration specified manually.

current_energy

The current energy level within the network

memory.matrix

A matrix of iteration, energy and node states (in columns, in that order), across all previous iterations (in rows).

stability_criterion

Criterion for stability. Changes below this value are no longer considered significant, and ten iterations without significant changes to the energy level in succession will trigger the convergence check.

output

Either 'qualitative' (default), in which case the check returns a boolean value representing whether it has passed or not, or 'quantitative', in which case the number of checked trials for which the convergence criterion was met is returned. This last option is of most value for debugging convergence.

Details

The check requires the following parameters: