Exposing numerical bugs in deep learning via gradient back-propagation-Reference-Cited by-同舟云学术

Exposing numerical bugs in deep learning via gradient back-propagation

Published:2021-08-18 Issue: Volume: Page:
ISSN:
Container-title:Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering
language:
Short-container-title:

Author:

Yan Ming¹,Chen Junjie¹^ORCID,Zhang Xiangyu²,Tan Lin²,Wang Gan¹,Wang Zan¹

Affiliation:

1. Tianjin University, China

2. Purdue University, USA

Funder

NSF (National Science Foundation)

ONR

National Natural Science Foundation of China

Intelligent Manufacturing Special Fund of Tianjin

IARPA TrojAI

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3468264.3468612

Reference56 articles.

1. Accessed: 2020. GitHub. https://github.com/philipperemy/deep-speaker/issues/5 Accessed: 2020. GitHub. https://github.com/philipperemy/deep-speaker/issues/5

2. Accessed: 2020. GitHub. https://github.com/ForeverZyh/TensorFlow-Program-Bugs/blob/master/StackOverflow/IPS-2/33699174-buggy/mnist.py Accessed: 2020. GitHub. https://github.com/ForeverZyh/TensorFlow-Program-Bugs/blob/master/StackOverflow/IPS-2/33699174-buggy/mnist.py

3. Accessed: 2020. GitHub. https://github.com/adamsolomou/SC-DNN/blob/a9169c6b7a0d456c1d2f229913e2d8c042c40aab/src/training/sc_train_creg.py Accessed: 2020. GitHub. https://github.com/adamsolomou/SC-DNN/blob/a9169c6b7a0d456c1d2f229913e2d8c042c40aab/src/training/sc_train_creg.py

4. Accessed: 2020. PyTorch Forums. https://discuss.pytorch.org/t/my-self-implemented-batchnorm-relu-gives-nan/42294 Accessed: 2020. PyTorch Forums. https://discuss.pytorch.org/t/my-self-implemented-batchnorm-relu-gives-nan/42294

5. Accessed: 2020. PyTorch Forums. https://discuss.pytorch.org/t/different-losses-on-2-different-machines/36446/5 Accessed: 2020. PyTorch Forums. https://discuss.pytorch.org/t/different-losses-on-2-different-machines/36446/5

Cited by 31 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. DualAttlog: Context aware dual attention networks for log-based anomaly detection;Neural Networks;2024-12

2. Large Language Models Can Connect the Dots: Exploring Model Optimization Bugs with Domain Knowledge-Aware Prompts;Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis;2024-09-11

3. Coverage-enhanced fault diagnosis for Deep Learning programs: A learning-based approach with hybrid metrics;Information and Software Technology;2024-09

4. A Miss Is as Good as A Mile: Metamorphic Testing for Deep Learning Operators;Proceedings of the ACM on Software Engineering;2024-07-12

5. Comparative analysis of real issues in open-source machine learning projects;Empirical Software Engineering;2024-05