Write a Blog >>
ESEC/FSE 2021
Mon 23 - Sat 28 August 2021 Athens, Greece

This program is tentative and subject to change.

Wed 25 Aug 2021 17:00 - 17:20 - AI—Software Engineering for Machine Learning 1
Thu 26 Aug 2021 05:00 - 05:20 - AI—Software Engineering for Machine Learning 1

There are increasing uses of deep learning (DL) compilers to generate optimized code, boosting the runtime performance of DL models on specific hardware. Like their traditional counterparts, DL compilers can generate incorrect code, resulting in unexpected model behaviors that may cause catastrophic consequences in mission-critical systems. On the other hand, the DL models processed by DL compilers differ fundamentally from imperative programs in that the program logic in DL models is implicit. As such, various characteristics of the bugs arising from traditional compilers need to be revisited in the context of DL compilers.

In this paper, we present the first systematic study of DL compiler bugs by analyzing 603 bugs arising in three popular DL compilers (i.e., TVM from Apache, Glow from Facebook, and nGraph from Intel). We analyzed these bugs according to their root causes, symptoms, and the stages where they occur during compilation. We obtain 12 findings, and provide a series of valuable guidelines for future work on DL compiler bug detection and debugging. For example, a large portion (nearly 20%) of DL compiler bugs are related to types, especially tensor types. The analysis of these bugs helps design new mutation operators (e.g., adding type cast for a tensor to promote implicit type conversion in subsequent tensor computations) to facilitate type-related bug detection. Further, we developed TVMfuzz as a proof-of-concept application of our findings to test the TVM DL compiler. It generates new tests based on TVM’s original test suite. They expose 8 TVM bugs that are missed by the original test suite. The result demonstrates the usefulness of our findings.

This program is tentative and subject to change.

Conference Day
Wed 25 Aug

Displayed time zone: Athens change

17:00 - 18:00
AI—Software Engineering for Machine Learning 1Research Papers +12h
17:00
20m
Research paper
A Comprehensive Study of Deep Learning Compiler Bugs
Research Papers
Qingchao ShenCollege of Intelligence and Computing, Tianjin University; School of New Media and Communication, Tianjin University, Haoyang MaCollege of Intelligence and Computing, Tianjin University, Junjie ChenTianjin University, Yongqiang TianUniversity of Waterloo, Shing-Chi CheungHong Kong University of Science and Technology, China, Xiang ChenNantong University
DOI
17:20
20m
Talk
Generating Efficient Solvers from Constraint Models
Research Papers
Shu LinPeking University, Na MengVirginia Tech, USA, Wenxin LiPeking University
17:40
20m
Talk
Probing Model Signal-Awareness via Prediction-Preserving Input Minimization
Research Papers
Yunhui ZhengIBM Research, Sahil SunejaIBM Research, Yufan ZhuangIBM Research, Jim A. LaredoIBM Research, USA, Alessandro MorariIBM Research

Conference Day
Thu 26 Aug

Displayed time zone: Athens change

05:00 - 06:00
AI—Software Engineering for Machine Learning 1Research Papers
05:00
20m
Research paper
A Comprehensive Study of Deep Learning Compiler Bugs
Research Papers
Qingchao ShenCollege of Intelligence and Computing, Tianjin University; School of New Media and Communication, Tianjin University, Haoyang MaCollege of Intelligence and Computing, Tianjin University, Junjie ChenTianjin University, Yongqiang TianUniversity of Waterloo, Shing-Chi CheungHong Kong University of Science and Technology, China, Xiang ChenNantong University
DOI
05:20
20m
Talk
Generating Efficient Solvers from Constraint Models
Research Papers
Shu LinPeking University, Na MengVirginia Tech, USA, Wenxin LiPeking University
05:40
20m
Talk
Probing Model Signal-Awareness via Prediction-Preserving Input Minimization
Research Papers
Yunhui ZhengIBM Research, Sahil SunejaIBM Research, Yufan ZhuangIBM Research, Jim A. LaredoIBM Research, USA, Alessandro MorariIBM Research