"Hierarchical testing designs for pattern recognition." Ann. This terminology highlights the fact that the families of analyses are examined sequen-tially and each one serves as a gatekeeper for the subsequent families. believe that the problem is not multiple testing but rather insufficient. Testing strategies considered here are commonly referred to as gatekeeping strategies. These ideas are illustrated in the context of detecting rectangles amidst clutter. Keywords: Bayesian inference, hierarchical modeling, multiple comparisons. In the assumptions ensuring this property a key role is played by the ratio cost/power. As might be expected, under mild assumptions good designs for sequential testing strategies exhibit a steady progression from broad scope coupled with low power to high power coupled with dedication to specific explanations. The total cost of a strategy is the sum of the “testing cost” and the “postprocessing cost” (proportional to | Ŷ|) and the corresponding optimization problem is analyzed. The set Ŷ is then taken to be the set of patterns that have not been ruled out by the tests performed. We consider sequential testing strategies in which decisions are made iteratively, based on past outcomes, about which test to perform next and when to stop testing. These tests are then characterized by scope (| A|), power (or type II error) and algorithmic cost. The focus here is then on pattern filtering: Given a large set $\mathcal $ are arranged in a hierarchy of nested partitions. Our formulation is motivated by applications to scene interpretation in which there are a great many possible explanations for the data, one (“background”) is statistically dominant, and it is imperative to restrict intensive computation to genuinely ambiguous regions. The object of the analysis is the computational process itself rather than probability distributions (Bayesian inference) or decision boundaries (statistical learning). We explore the theoretical foundations of a “twenty questions” approach to pattern recognition.
0 Comments
Leave a Reply. |