AdaStop: Cost-Aware Early Stopping for DNN Test Selection

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 $\mathit{\text{AdaStop}}$, a framework that estimates the marginal fault discovery rate during testing and stops labeling when the estimated rate falls below the threshold $\tau =c\mathrm{/}v$. Experiments across multiple datasets, architectures, and selection strategies show that $65$–$84\mathrm{\%}$ of faults can be discovered using only $9$–$31\mathrm{\%}$ of the labeling budget.