arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.09314cs.SE
收藏

Themis:具备强故障检测能力的自动高效深度学习系统测试方法

Themis: Automatic and Efficient Deep Learning System Testing with Strong Fault Detection Capability

Dong Huang, Tsz On Li, Xiaofei Xie, Heming Cui

AI总结:

针对深度学习系统测试需手动指定神经元激活值、故障检测率低的问题,提出自动测试系统Themis,通过自动挖掘扰动下神经元输出变化显著的致错数据流实现高覆盖检测,实验证明其故障检出量和重训练后的准确率提升均大幅优于现有方法。

摘要:

深度学习系统(DLS)已广泛应用于自动驾驶等安全关键任务中。然而,当向DLS输入受扰动的样本进行推理时,系统常会产生错误输出(即故障)。DeepXplore等DLS测试技术通过生成扰动输入,探索会诱发故障的数据流来检测此类问题。由于DLS通常存在无穷多的数据流,现有技术需要开发者手动指定DLS神经元的一组激活值,以探索致错数据流。但近期研究表明,这种人工操作繁琐,且仅能检测到极少比例的致错数据流。\n本文提出Themis——首个自动DLS测试系统,它通过高概率全面覆盖致错数据流,实现了强大的故障检测能力。Themis采用全新工作流,可自动、系统地挖掘这类数据流:当输入发生轻微扰动时,其内部神经元输出会出现显著变化,这类数据流大概率会诱发故障。我们在10个不同的DLS上对Themis进行了评估,结果显示,Themis检测到的故障数量平均是4种主流DLS测试技术的3.78倍。利用检测到的故障对所有受测DLS进行重训练后,Themis还使这些系统的准确率提升(恢复)幅度平均比所有基线方法高14.7倍。

英文摘要:

Deep Learning Systems (DLSs) have been widely applied in safety-critical tasks such as autopilot. However, when a perturbed input is fed into a DLS for inference, the DLS often has incorrect outputs (i.e., faults). DLS testing techniques (e.g., DeepXplore) detect such faults by generating perturbed inputs to explore data flows that induce faults. Since a DLS often has infinitely many data flows, existing techniques require developers to manually specify a set of activation values in a DLS's neurons for exploring fault-inducing data flows. Unfortunately, recent studies show that such manual effort is tedious and can detect only a tiny proportion of fault-inducing data flows. In this paper, we present Themis, the first automatic DLS testing system, which attains strong fault detection capability by ensuring a full coverage of fault-inducing data flows at a high probability. Themis carries a new workflow for automatically and systematically revealing data flows whose internal neurons' outputs vary substantially when the inputs are slightly perturbed, as these data flows are likely fault-inducing. We evaluated Themis on ten different DLSs and found that on average the number of faults detected by Themis was 3.78X more than four notable DLS testing techniques. By retraining all evaluated DLSs with the detected faults, Themis also increased (regained) these DLSs' accuracies on average 14.7X higher than all baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑