The Animal-AI Environment: Training and Testing Animal-Like Artificial Cognition

TL;DR

This paper introduces the Animal-AI Environment, a game-based platform designed to evaluate artificial agents' understanding and modeling of their environment, inspired by animal cognition research.

Contribution

It presents a novel environment that combines the versatility of game settings with explicit testing of animal-like cognition in AI agents.

Findings

Environment effectively tests cognitive understanding

Agents demonstrate varied levels of problem-solving skills

Platform facilitates future research in animal-inspired AI

Abstract

Recent advances in artificial intelligence have been strongly driven by the use of game environments for training and evaluating agents. Games are often accessible and versatile, with well-defined state-transitions and goals allowing for intensive training and experimentation. However, agents trained in a particular environment are usually tested on the same or slightly varied distributions, and solutions do not necessarily imply any understanding. If we want AI systems that can model and understand their environment, we need environments that explicitly test for this. Inspired by the extensive literature on animal cognition, we present an environment that keeps all the positive elements of standard gaming environments, but is explicitly designed for the testing of animal-like artificial cognition.