Wains: A Pattern-Seeking Artificial Life Species

Abstract

We describe the initial phase of a research project to develop an artificial life framework designed to extract knowledge from large data sets with minimal preparation or ramp-up time. In this phase, we evolved an artificial life population with a new brain architecture. The agents have sufficient intelligence to discover patterns in data and to make survival decisions based on those patterns. The species uses diploid reproduction, Hebbian learning, and Kohonen self-organizing maps, in combination with novel techniques such as using pattern-rich data as the environment and framing the data analysis as a survival problem for artificial life. The first generation of agents mastered the pattern discovery task well enough to thrive. Evolution further adapted the agents to their environment by making them a little more pessimistic, and also by making their brains more efficient.

Evaluation of wains as a classifier for automatic speech recognition

“…This will be done by making the training environment harsher. As proven in [14], wains are able to adapt to their environment. By making the environment harsher, the wains will have a greater incentive to learn and achieve better recognition performance.…”

“…The framework used to develop the new back-end classifier is Créatúr [12]. Créatúr is a reusable framework supporting the creation of artificial life forms and can be used in two ways.…”

“…This tutorial outlines how to create ALife species using Créatúr. In [12], Créatúr was used to develop an ALife species called wains that was initially used to discover patterns in handwritten numerals.…”

A Life-Based Classifier for Automatic Speech Recognition

“…The Créatúr Framework. Créatúr [12] is a reusable framework that automates experiments with artificial life forms and encapsulates functionality common to most ALife species, such as eating, mating, and playing; users have the ability to implement additional features as needed. The framework can be used in two ways: as a tool for analysing complex data and discovering patterns or as a tool for studying ALife [12].…”

A Functional Approach to Sex: Reproduction in the Créatúr Framework

“…Are there tools that can make this easier? As part of our research using ALife to extract knowledge from large data sets with minimal preparation or ramp-up time [8], we work with a variety of agents. We developed Créatúr 1 , which is both a software framework for automating experiments with ALife and a library of modules that can be used (with or without the framework) to implement agents.…”