paper-with-me

홈 › Papers

AdaStop: Cost-Aware Early Stopping for DNN Test Selection

2026-07-06 · Bonan Shen, Wei-Jung Huang, Xin Liu, Jiazhou Gao, Tao Ning arxiv

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.

📄 PDF Abstract BibTeX arXiv:2607.05461

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

AdaStop: adaptive statistical testing for sound comparisons of Deep RL agents

2023-06-19 · Timothée Mathieu, Riccardo Della Vecchia, Alena Shilova, Matheus Medeiros Centa 외

Recently, the scientific community has questioned the statistical reproducibility of many empirical results, especially in the field of machine learning. To contribute to the resolution of this reproducibility crisis, we…

Deep Reinforcement LearningMuJoCoReinforcement Learning (RL)

ACE: Adaptive Constraint-aware Early Stopping in Hyperparameter Optimization

2022-08-04 · Yi-Wei Chen, Chi Wang, Amin Saied, Rui Zhuang

Deploying machine learning models requires high model quality and needs to comply with application constraints. That motivates hyperparameter optimization (HPO) to tune model configurations under deployment constraints. …

FairnessHyperparameter Optimization

Dynamic Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Sampling

2024-08-30 · Guangya Wan, Yuqi Wu, Jie Chen, Sheng Li

Self-Consistency (SC) is a widely used method to mitigate hallucinations in Large Language Models (LLMs) by sampling the LLM multiple times and outputting the most frequent solution. Despite its benefits, SC results in s…

Statistical Early Stopping for Reasoning Models

2026-02-15 · Yangxinyu Xie, Tao Wang, Soham Mallick, Yan Sun 외 arxiv

While LLMs have seen substantial improvement in reasoning capabilities, they also sometimes overthink, generating unnecessary reasoning steps, particularly under uncertainty, given ill-posed or ambiguous queries. We intr…

FLOP-Efficient Training: Early Stopping Based on Test-Time Compute Awareness

2026-01-04 · Hossam Amer, Maryam Dialameh, Hossein Rajabzadeh, Walid Ahmed 외 arxiv

Scaling training compute, measured in FLOPs, has long been shown to improve the accuracy of large language models, yet training remains resource-intensive. Prior work shows that increasing test-time compute (TTC)-for exa…