Proceedings of the 44th International Conference on Software Engineering · 2022 · 64 citations · 41 references
ComputingComputer EngineeringComputer ScienceKnowledge RepresentationMachine LearningSoftware EngineeringDeep LearningFuzz TestingNeural NetworksRepresentation LearningSoftware TestingSoftware Libraries
Deep learning (DL) techniques are proven effective in many challenging tasks, and become widely-adopted in practice. However, previous work has shown that DL libraries, the basis of building and executing DL models, contain bugs and can cause severe consequences. Unfortunately, existing testing approaches still cannot comprehensively exercise DL libraries. They utilize existing trained models and only detect bugs in model inference phase. In this work we propose Muffin to address these issues. To this end, Muffin applies a specifically-designed model fuzzing approach, which allows it to generate diverse DL models to explore the target library, instead of relying only on existing trained models. Muffin makes differential testing feasible in the model training phase by tailoring a set of metrics to measure the inconsistencies between different DL libraries. In this way, Muffin can best exercise the library code to detect more bugs. To evaluate the effectiveness of Muffin, we conduct experiments on three widely-used DL libraries. The results demonstrate that Muffin can detect 39 new bugs in the latest release versions of popular DL libraries, including Tensorflow, CNTK, and Theano.
41
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. · 2016 · 222.1K citations · Full text
Densely Connected Convolutional Networks
Gao Huang, Zhuang Liu, Laurens van der Maaten et al. · 2017 · 44.7K citations
Recurrent neural network based language model
Tomáš Mikolov, Martin Karafiát, Lukáš Burget et al. · 2010 · 5.4K citations
Neural Architecture Search: A Survey
Thomas Elsken, Jan Hendrik Metzen, Frank Hutter · arXiv (Cornell University) · 2018 · 1.4K citations · Full text
Kexin Pei, Yinzhi Cao, Junfeng Yang et al. · 2017 · 1.2K citations · Full text