Mindmodeling Home. . . and Anywhere Else You Have Idle Processors

摘要

As we, the ICCM community, continue to expand the scope of our cognitive modeling ambitions, we increasingly face computational requirements that are an impediment to progress. Computational complexity grows quickly with increases in the granularity of models, the fidelity of the models' operating environment, and the time scales across which these models interact. Additional processing demands are encountered when studying the breadth of a cognitive model's performance capabilities such as through observing the model's sustained fitness while varying the environment or conducting sensitivity analyses of interactions between internal model parameters in a controlled experiments. Such computational demands are not unique to the cognitive modeling community. Other scientific fields (bioinformatics, meteorology, physics, etc.) have already pioneered a variety of platforms and methodologies for dealing with similarly computationally complex problems. We will achieve faster progress toward the broader scientific objectives of cognitive modeling and the specific goals of particular research projects if we pay attention to the lessons learned and capabilities developed in other computational sciences.