Recent successes in Artificial Intelligence (AI) use machine learning to produce AI agents with both hand-engineered and procedurally generated elements learned from large amounts of data. As the balance shifts toward procedural generation, how can we predict interactions between such agents and humans? We propose to use Artificial Life to study emergence of group behaviours between procedurally generated AI agents and humans. We simulate Darwinian evolution to procedurally generate agents in a simple environment where the agents interact with human-controlled avatars. To reduce human involvement time, we machine-learn another set of AI agents that mimic human avatar behaviours and run the evolution with such human proxies instead of actual humans. This paper is an update on the on-going project.

Heuristic search is a core area of Artificial Intelligence, successfully applied to planning, constraint satisfaction and game playing. In real-time heuristic search autonomous agents interleave planning and plan execution and access environment locally which make them more suitable for Artificial Life style settings. Over the last two decades a large number of real-time heuristic search algorithms have been manually crafted and evaluated. In this paper we break down several published algorithms into building blocks and then let a simulated evolution re-combine the blocks in a performance-based way. Remarkably, even relatively short evolution runs result in algorithms with state-of-the-art performance. These promising preliminary results open exciting possibilities in the field of real-time heuristic search.