Qualitative Probabilistic Networks for Planning Under Uncertainty
This work addresses planning under uncertainty for AI systems, but it is incremental as it builds on Bayesian networks with qualitative reasoning.
The paper tackles the problem of planning under uncertainty by introducing qualitative probabilistic networks, which use a probabilistic semantics for qualitative likelihood assertions to derive conclusions about action influences, offering weaker but valuable insights for suggesting actions, eliminating inferior plans, identifying tradeoffs, and explaining models.
Bayesian networks provide a probabilistic semantics for qualitative assertions about likelihood. A qualitative reasoner based on an algebra over these assertions can derive further conclusions about the influence of actions. While the conclusions are much weaker than those computed from complete probability distributions, they are still valuable for suggesting potential actions, eliminating obviously inferior plans, identifying important tradeoffs, and explaining probabilistic models.