一句话：通过大规模实证研究并提出基于约束的 ShapeTracer，检测工业 TensorFlow 程序的形状相关缺陷。
深度学习在工业界广泛应用，但程序缺陷可能导致严重故障与资源浪费。本文针对工业 TensorFlow 程序进行缺陷检测，首先对 12,289 个失败的 TensorFlow 作业开展实证研究，发现现有静态工具可有效检测排名前三的 Python 缺陷类型，检测率为 72.55%。针对 TensorFlow 特有的形状相关错误，作者首次提出基于约束的检测方法并实现工具 ShapeTracer。该工具在 60 个工业 TensorFlow 程序上评估：每个程序分析不超过 3 秒，成功检测 40 个程序缺陷且无假阳性。ShapeTracer 已部署于 platform-X 平台，并将开源发布。
Deep learning has been widely adopted in industry and has achieved great success in a wide range of application areas. Bugs in deep learning programs can cause catastrophic failures, in addition to a serious waste of resources and time.This paper aims at detecting industrial TensorFlow program bugs. We report an extensive empirical study on 12,289 failed TensorFlow jobs, showing that existing static tools can effectively detect 72.55% of the top three types of Python bugs in industrial TensorFlow programs. In addition, we propose (for the first time) a constraint-based approach for detecting TensorFlow shape-related errors (one of the most common TensorFlow-specific bugs), together with an associated tool, ShapeTracer. Our evaluation on a set of 60 industrial TensorFlow programs shows that ShapeTracer is efficient and effective: it analyzes each program in at most 3 seconds and detects effectively 40 out of 60 industrial TensorFlow program bugs, with no false positives. ShapeTracer has been deployed in the platform-X platform and will be released soon.
DOI 原文 ·
@inproceedings{paperbot6618,
title = {Detecting TensorFlow Program Bugs in Real-World Industrial Environment},
author = {Chen Liu and Jie Lu and Guangwei Li and Ting Yuan and Lian Li and Feng Tan and Jun Yang and Liang You and Jingling Xue},
booktitle = {ASE},
year = {2021},
doi = {10.1109/ase51524.2021.9678891}
}