Towards robust models of code via energy-based learning on auxiliary datasets
Existing approaches to improving the robustness of source code models concentrate on recognizing adversarial samples rather than valid samples that fall outside of a given distribution, which we refer to as out-of-distribution (OOD) samples. To this end, we propose to use an auxiliary dataset (out-o...
Saved in:
Main Authors: | BUI, Duy Quoc Nghi, YU, Yijun |
---|---|
格式: | text |
語言: | English |
出版: |
Institutional Knowledge at Singapore Management University
2022
|
主題: | |
在線閱讀: | https://ink.library.smu.edu.sg/sis_research/10117 https://ink.library.smu.edu.sg/context/sis_research/article/11117/viewcontent/RobustModelsCode_pv.pdf |
標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
相似書籍
-
Self-supervised contrastive learning for code retrieval and summarization via semantic-preserving transformations
由: BUI, Duy Quoc Nghi, et al.
出版: (2021) -
InferCode: Self-supervised learning of code representations by predicting subtrees
由: BUI, Duy Quoc Nghi, et al.
出版: (2021) -
TreeCaps: Tree-based capsule networks for source code processing
由: BUI, Duy Quoc Nghi, et al.
出版: (2021) -
AutoFocus: Interpreting attention-based neural networks by code perturbation
由: BUI, Duy Quoc Nghi, et al.
出版: (2019) -
Novel deep learning methods combined with static analysis for source code processing
由: BUI, Duy Quoc Nghi
出版: (2020)