Path Analysis Models of an Autonomous Agent in a Complex Environment

We seek explanatory models of how and why AI systems work in particular environments. We are not satisfied to demonstrate performance, we want to understand it. In terms of data and models, this means we are not satisfied with descriptive summaries, nor even with predictive models. We want causal models. In this brief abstract we will present descriptive, predictive and causal models of the behavior of agents that fight simulated forest fires. We will describe the shortcomings of descriptive and predictive models, and summarize path analysis, a common technique for inducing causal models.