AdaStop: Cost-Aware Early Stopping for DNN Test Selection
It solves the budget selection problem in DNN testing for practitioners, offering a principled early stopping method.
AdaStop formulates DNN testing as a cost-benefit decision process and stops labeling when the marginal fault discovery rate falls below a threshold, discovering 65-84% of faults using only 9-31% of the labeling budget.
Existing methods for testing deep neural networks (DNNs) primarily prioritize test inputs likely to reveal model faults under a fixed labeling budget. In practice, choosing that budget is difficult: too little testing misses failures, while too much incurs unnecessary labeling costs. This work studies the stopping problem in DNN testing. We formulate testing as a cost--benefit decision process in which labeling an input incurs cost $c$ and discovering a fault yields value $v$. Based on this formulation, we introduce \textit{AdaStop}, a framework that estimates the marginal fault discovery rate during testing and stops labeling when the estimated rate falls below the threshold $τ= c/v$. Experiments across multiple datasets, architectures, and selection strategies show that $65$--$84\%$ of faults can be discovered using only $9$--$31\%$ of the labeling budget.