Functional Equivalence with NARS

TL;DR

The paper investigates functional equivalence within the Non-Axiomatic Reasoning System (NARS) by modifying OpenNARS for Applications (ONA) to derive higher-order contingencies and transfer knowledge across modalities. It presents an extended, stepwise learning framework that enables reading-like tasks and cross-modal transfer (e.g., objects to words and words to objects) through identity matching, auditory comprehension, and reading comprehension. Key contributions include formalizing functional equivalence as a derived relation, demonstrating transfer learning across modalities, and showing how a NARS-based system can ground language-building blocks via structured training. The work advances cognitive flexibility in AGI, with potential practical impact on language-grounded reasoning and autonomous robotics in dynamic environments.

Abstract

This study explores the concept of functional equivalence within the framework of the Non-Axiomatic Reasoning System (NARS), specifically through OpenNARS for Applications (ONA). Functional equivalence allows organisms to categorize and respond to varied stimuli based on their utility rather than perceptual similarity, thus enhancing cognitive efficiency and adaptability. In this study, ONA was modified to allow the derivation of functional equivalence. This paper provides practical examples of the capability of ONA to apply learned knowledge across different functional situations, demonstrating its utility in complex problem-solving and decision-making. An extended example is included, where training of ONA aimed to learn basic human-like language abilities, using a systematic procedure in relating spoken words, objects and written words. The research carried out as part of this study extends the understanding of functional equivalence in AGI systems, and argues for its necessity for level of flexibility in learning and adapting necessary for human-level AGI.

Figures (1)

• Figure 1: The six step involved in the example.