Oct 30, 2023 · Abstract:Static and dynamic computational graphs represent two distinct approaches to constructing deep learning frameworks.

Nov 21, 2023 · Fuzzing has been applied to deep learning systems to automatically detect bugs and improve their robustness. CRADLE [45] leverages existing DL ...

Nov 26, 2023 · Static and dynamic computational graphs represent two distinct approaches to constructing deep learning frameworks.

This work proposes several code transformations to generate test cases involving dynamic features and successfully identified twenty previously unknown bugs ...

While recent research has introduced fuzzing to test deep learning compilers, there is still a lack of comprehensive analysis on how to test dynamic features.

While recent research has introduced fuzzing to test deep learning compilers, there is still a lack of comprehensive analysis on how to test dynamic features.

TorchProbe: Fuzzing Dynamic Deep Learning Compilers. https://doi.org/10.1007/978-981-99-8311-7_15 ·. Journal: Programming Languages and Systems Lecture Notes ...

We propose HirGen, an automated testing technique that effectively exposes coding mistakes in the optimization of high-level IRs.

Oct 30, 2023 · Static and dynamic computational graphs represent two distinct approaches to constructing deep learning frameworks.

Oct 17, 2024 · TorchProbe: Fuzzing Dynamic Deep Learning Compilers. Qidong Su, Chuqin Geng, Gennady Pekhimenko, Xujie Si. APLAS 2023. Adaptive Load Balancing ...