Shallow variational bottleneck compression is more vulnerable to adversarial attacks than deep variational bottleneck compression, and generative decoders widen the attack surface in task-oriented communication.
Frankensplit: Efficient neural feature compression with shallow variational bottleneck injection for mobile edge comput- ing,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Adversarial Robustness of Bottleneck Injected Deep Neural Networks for Task-Oriented Communication
Shallow variational bottleneck compression is more vulnerable to adversarial attacks than deep variational bottleneck compression, and generative decoders widen the attack surface in task-oriented communication.