On the influence of intelligence in (social) intelligence testing environments

TL;DR

This paper investigates how varying levels of agent intelligence affect social intelligence testing environments, aiming to improve evaluation methods through analysis of reinforcement algorithms in cooperative and competitive contexts.

Contribution

It introduces a framework for analyzing the impact of agent intelligence levels on social intelligence assessments using reinforcement learning in multiagent systems.

Findings

Different reinforcement algorithms show varied effectiveness in social intelligence testing.

Agent cooperation and competition dynamics influence test outcomes.

The Darwin-Wallace inspired distribution provides a novel context for analysis.

Abstract

This paper analyses the influence of including agents of different degrees of intelligence in a multiagent system. The goal is to better understand how we can develop intelligence tests that can evaluate social intelligence. We analyse several reinforcement algorithms in several contexts of cooperation and competition. Our experimental setting is inspired by the recently developed Darwin-Wallace distribution.