Neural network semantic backdoor detection and mitigation: A causality-based approach
Different from ordinary backdoors in neural networks which are introduced with artificial triggers (e.g., certain specific patch) and/or by tampering the samples, semantic backdoors are introduced by simply manipulating the semantic, e.g., by labeling green cars as frogs in the training set. By focu...
Saved in:
Main Authors: | SUN, Bing, SUN, Jun, KOH, Wayne, SHI, Jie |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2024
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/9211 https://ink.library.smu.edu.sg/context/sis_research/article/10217/viewcontent/sec23winter_prepub_118_sun.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
Similar Items
-
Certified continual learning for neural network regression
by: PHAM, Hong Long, et al.
Published: (2024) -
QEBVerif: Quantization error bound verification of neural networks
by: ZHANG, Yedi, et al.
Published: (2023) -
Decision-guided weighted automata extraction from recurrent neural networks
by: ZHANG, Xiyue, et al.
Published: (2021) -
Watch out! Motion is blurring the vision of your deep neural networks
by: GUO, Qing, et al.
Published: (2020) -
A quantitative analysis framework for recurrent neural network
by: DU, Xiaoning, et al.
Published: (2019)