Evolving Self-taught Neural Networks: The Baldwin Effect and the Emergence of Intelligence

TL;DR

This paper introduces evolving self-taught neural networks that combine evolution and intrinsic motivation, leading to emergent intelligent foraging strategies in autonomous agents.

Contribution

It presents a novel method where neural networks self-teach without external rewards, demonstrating the interplay of evolution and learning in developing intelligence.

Findings

Self-taught neural networks outperform purely evolutionary or self-teaching approaches.

Interaction between evolution and self-teaching leads to emergent intelligent behaviors.

Experimental results show improved foraging strategies in autonomous agents.

Abstract

The so-called Baldwin Effect generally says how learning, as a form of ontogenetic adaptation, can influence the process of phylogenetic adaptation, or evolution. This idea has also been taken into computation in which evolution and learning are used as computational metaphors, including evolving neural networks. This paper presents a technique called evolving self-taught neural networks - neural networks that can teach themselves without external supervision or reward. The self-taught neural network is intrinsically motivated. Moreover, the self-taught neural network is the product of the interplay between evolution and learning. We simulate a multi-agent system in which neural networks are used to control autonomous agents. These agents have to forage for resources and compete for their own survival. Experimental results show that the interaction between evolution and the ability to teach oneself in self-taught neural networks outperform evolution and self-teaching alone. More specifically, the emergence of an intelligent foraging strategy is also demonstrated through that interaction. Indications for future work on evolving neural networks are also presented.