Proof of correctness for ASOCS AA3 networks

Abstract

Simplified generation of all mechanical assembly sequences," IEEE J.Abstract-This paper analyzes adaptive algorithm 3 (AA3) of adaptive self-organizing concurrent systems (ASOCS) and proves that AA3 correctly fulfills the rules presented. Several different models for ASOCS have been developed. AA3 uses a distributed mechanism for implementing rules so correctness is not obvious. An ASOCS is an adaptive network composed of many simple computing elements operating in parallel. An ASOCS operates in one of two modes: learning and processing. In learning mode, rules are presented to the ASOCS and incorporated in a self-organizing fashion. In processing mode, the ASOCS acts as a parallel hardware circuit that performs the function defined by the learned rules.

Learning by Discrimination: A Constructive Incremental Approach

“…Several learning algorithms have been developed for ASOCS, including Adaptive Algorithm (AA)1 [24], AA2 [26] and AA3 [27]. Although all three algorithms have been shown to converge on any arbitrary Boolean pattern set, only AA1 exhibits meaningful off-training set generalization [4], [13].…”

A Constructive Incremental Learning Algorithm for Binary Classification Tasks

“…Several learning algorithms have been developed for ASOCS, including Adaptive Algorithm (AA)1 [14], AA2 [15] and AA3 [16]. Although all three algorithms have been shown to converge on any arbitrary Boolean pattern set, only AA1 exhibits meaningful out-of-sample generalization [17], [18].…”