Zobrazeno 1 - 10
of 19
pro vyhledávání: '"Fu, Zhoulai"'
With the increasing complexity of machine learning models, managing computational resources like memory and processing power has become a critical concern. Mixed precision techniques, which leverage different numerical precisions during model trainin
Externí odkaz:
http://arxiv.org/abs/2305.10947
Lowering the precision of neural networks from the prevalent 32-bit precision has long been considered harmful to performance, despite the gain in space and time. Many works propose various techniques to implement half-precision neural networks, but
Externí odkaz:
http://arxiv.org/abs/2301.12809
Autor:
Su, Ting, Zhang, Chengyu, Yan, Yichen, Fan, Lingling, Pu, Geguang, Liu, Yang, Fu, Zhoulai, Su, Zhendong
Data-flow testing (DFT) aims to detect potential data interaction anomalies by focusing on the points at which variables receive values and the points at which these values are used. Such test objectives are referred as \emph{def-use pairs}. However,
Externí odkaz:
http://arxiv.org/abs/1803.10431
Autor:
Fu, Zhoulai, Su, Zhendong
Achieving high code coverage is essential in testing, which gives us confidence in code quality. Testing floating-point code usually requires painstaking efforts in handling floating-point constraints, e.g., in symbolic execution. This paper turns th
Externí odkaz:
http://arxiv.org/abs/1704.03394
Autor:
Fu, Zhoulai, Su, Zhendong
This paper presents Mathematical Execution (ME), a new, unified approach for testing numerical code. The key idea is to (1) capture the desired testing objective via a representing function and (2) transform the automated testing problem to the minim
Externí odkaz:
http://arxiv.org/abs/1610.01133
Autor:
Fu, Zhoulai
Si la production de logiciel fiable est depuis longtemps la préoccupation d'ingénieurs, elle devient à ce jour une branche de sujets de recherche riche en applications, dont l'analyse statique. Ce travail a porté sur l'analyse statique de program
Externí odkaz:
http://www.theses.fr/2013REN1S060/document
Reducing the number of bits needed to encode the weights and activations of neural networks is highly desirable as it speeds up their training and inference time while reducing memory consumption. For these reasons, research in this area has attracte
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a2e965cbffbe60f1b34b40d6bb1e9255
http://arxiv.org/abs/2305.10947
http://arxiv.org/abs/2305.10947
Autor:
Fu, Zhoulai
Publikováno v:
Other [cs.OH]. Université Rennes 1, 2013. English. ⟨NNT : 2013REN1S060⟩
Other [cs.OH]. Université de Rennes; Université européenne de Bretagne (2007-2016), 2013. English. ⟨NNT : 2013REN1S060⟩
Other [cs.OH]. Université de Rennes; Université européenne de Bretagne (2007-2016), 2013. English. ⟨NNT : 2013REN1S060⟩
The fast and furious pace of change in computing technology has become an article of faith for many. The reliability of computer-based systems cru- cially depends on the correctness of its computing. Can man, who created the computer, be capable of p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::cf157bec7d8fa41ab966e378555218dc
https://tel.archives-ouvertes.fr/tel-00918593/file/FU_Zhoulai.pdf
https://tel.archives-ouvertes.fr/tel-00918593/file/FU_Zhoulai.pdf
Autor:
Fu, Zhoulai, Su, Zhendong
Publikováno v:
Computer Aided Verification: 28th International Conference, CAV 2016, Toronto, ON, Canada, July 17-23, 2016, Proceedings, Part II; 2016, p187-209, 23p
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.