The hidden dangers of experimenting in distributed AI

@inproceedings{Smit2006TheHD,
  title={The hidden dangers of experimenting in distributed AI},
  author={Annika Smit and Masja Kempen and Niek J. E. Wijngaards and Andr{\'e} Meyer},
  booktitle={AAMAS},
  year={2006}
}
Research on multi-agent systems often involves experiments, also in situations where humans interact with agents. Consequently, the field of experimental (human) sciences becomes more and more relevant. This paper clarifies how things can and often do go wrong in distributed AI experiments. We show the flaws in methodological design in existing papers (both… CONTINUE READING