PUBLIC MARKS with tags 2005 & gameai

Gamespot's Best AI of 2005

"Forming expectations is a problem that is relatively easy to concretize: AIs have an internal knowledge model of the world. If the AI is able to provide an estimate for what that state is going to look like n ticks in the future, that’s an expectation – and naturally we’re going to want the AI to react not (or not just) to current world-state, but also to its expectation. React to its own expectation. I think that’s a neat idea, and architecturally it makes a lot of sense. Assuming we have a reactive behavior system of some sort already, we don’t, as a first pass, need to modify IT at all. All we need to do is provide future instead of current data. Great!"

In this paper, an implementation of "Stackless Python" (a Python which does not keep state on the C stack) is presented. Surprisingly, the necessary changes affect just a small number of C modules, and a major rewrite of the C library can be avoided. The key idea in this approach is a paradigm change for the Python code interpreter that is not easy to understand in the first place. Recursive interpreter calls are turned into tail recursion, which allows deferring evaluation by pushing frames to the frame stack, without the C stack involved.