.png){kind=icon}
.png){kind=icon}
Return
FSWF-APMA: Few-Shot Website Fingerprinting with Autoencoder and Prototype Matching Alignment
.png){kind=icon}
PRE
AI
Y
W
Q
X
J
DOI：10.1016/j.patrec.2026.05.005.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• The proposed method achieves over 93 shot scenarios. • Autoencoder pre-trains features from unlabeled Tor fingerprint data. • Prototype matching loss prevents classifier overfitting on tasks. • Ablation experiments proves the effectiveness of the component.
Keywords:
Few-shot learning
Website fingerprinting
Autoencoder
Prototype matching
Feature alignment
Journal
IF:
3.3
Papers:
7.8K
Citations:
1.6W
Organization
Cited Papers