Zuoyan Zhang

Zuoyan Zhang is currently a first-year PhD student at the CYCLE Lab at Hunan University under the supervisor Prof. Jie Zhao. He got MEng degree from the State Key Laboratory of Mathematical Engineering and Advanced Computing at Information Engineering University under the supervisor Prof. Shaozhong Guo and Assoc. Prof. Jinchen Xu in June 2024. His research interests span machine learning system and numerical program analysis, with a particular focus on floating-point program error detection and automatic repair, as well as distributed parallel acceleration for large language model training. In the area of distributed training acceleration for LLM, his research focuses on automatic parallel strategy search and communication-hiding. His research has been recognized with ASE, ISSTA.

news

Sep 26, 2024	I participated in CCF HPC China 2024 and be a guest speaker at the 2nd China Chip Basic Function Library Forum.
Sep 19, 2024	I participated in ISSTA 2024 and gave a presentation on “Arfa: An Agile Regime-based Floating-point Optimization Approach for Rounding Errors”.
Jul 03, 2024	One paper titled “Arfa: An Agile Regime-based Floating-point Optimization Approach for Rounding Errors” was accepted by ISSTA 2024.
Jul 11, 2023	One paper titled “Eiffel: Inferring Input Ranges of Significant Floating-point Errors via Polynomial Extrapolation” was accepted by ASE 2023.

selected publications

ASE 2023
Eiffel: Inferring Input Ranges of Significant Floating-Point Errors via Polynomial Extrapolation

Zuoyan Zhang, Bei Zhou, Jiangwei Hao, Hongru Yang, and 6 more authors

In Proceedings of the 38th IEEE/ACM International Conference on Automated Software Engineering, Echternach, Luxembourg, 2023

Abs DOI Bib PDF Video Slides

Existing search heuristics used to find input values that result in significant floating-point (FP) errors or small ranges that cover them are accompanied by severe constraints, complicating their implementation and restricting their general applicability. This paper introduces an error analysis tool called Eiffel to infer error-inducing input ranges instead of searching them. Given an FP expression with its domain D, Eiffel first constructs an error data set by sampling values across a smaller domain ℛ and assembles these data into clusters. If more than two clusters are formed, Eiffel derives polynomial curves that best fit the bound coordinates of the error-inducing ranges in ℛ, extrapolating them to infer all target ranges of D and reporting the maximal error. Otherwise, Eiffel simply returns the largest error across ℛ. Experimental results show that Eiffel exhibits a broader applicability than Atomu and S3FP by successfully detecting the errors of all 70 considered benchmarks while the two baselines only report errors for part of them. By taking as input the inferred ranges of Eiffel, Herbie obtains an average accuracy improvement of 3.35 bits and up to 53.3 bits.
@inproceedings{ase 2023, author = {Zhang, Zuoyan and Zhou, Bei and Hao, Jiangwei and Yang, Hongru and Cui, Mengqi and Zhou, Yuchang and Song, Guanghui and Li, Fei and Xu, Jinchen and Zhao, Jie}, title = {Eiffel: Inferring Input Ranges of Significant Floating-Point Errors via Polynomial Extrapolation}, year = {2023}, isbn = {9798350329964}, publisher = {IEEE Press}, url = {https://doi.org/10.1109/ASE56229.2023.00139}, doi = {10.1109/ASE56229.2023.00139}, booktitle = {Proceedings of the 38th IEEE/ACM International Conference on Automated Software Engineering}, pages = {1441–1453}, numpages = {13}, location = {Echternach, Luxembourg}, dimensions = {true}, series = {ASE '23} }