The Automatic Programming of Agents that Learn Mental Models and Create Simple Plans of Action

An essential component of an intelligent agent is the ability to notice, encode, store, and utilize information about its environment. Traditional approaches to program induction have focused on evolving functional or reactive programs. This paper presents MAPMAKER, an approach to the automatic generation of agents that discover information about their environment, encode this information for later use, and create simple plans utilizing the stored mental models. In this approach, agents are multi-part computer programs that communicate through a shared memory. Both the programs and the representation scheme are evolved using genetic programming. An illustrative problem of 'gold' collection is used to demonstrate the approach in which one part of a program makes a map of the world and stores it in memory, and the other part uses this map to find the gold The results indicate that the approach can evolve programs that store simple representations of their environments and use these representations to produce simple plans.