High-Robustness, Low-Transferability Fingerprinting of Neural Networks

05/14/2021
by   Siyue Wang, et al.
0

This paper proposes Characteristic Examples for effectively fingerprinting deep neural networks, featuring high-robustness to the base model against model pruning as well as low-transferability to unassociated models. This is the first work taking both robustness and transferability into consideration for generating realistic fingerprints, whereas current methods lack practical assumptions and may incur large false positive rates. To achieve better trade-off between robustness and transferability, we propose three kinds of characteristic examples: vanilla C-examples, RC-examples, and LTRC-example, to derive fingerprints from the original base model. To fairly characterize the trade-off between robustness and transferability, we propose Uniqueness Score, a comprehensive metric that measures the difference between robustness and transferability, which also serves as an indicator to the false alarm problem.

READ FULL TEXT
research
09/19/2022

On the Adversarial Transferability of ConvMixer Models

Deep neural networks (DNNs) are well known to be vulnerable to adversari...
research
09/07/2022

On the Transferability of Adversarial Examples between Encrypted Models

Deep neural networks (DNNs) are well known to be vulnerable to adversari...
research
04/01/2021

TRS: Transferability Reduced Ensemble via Encouraging Gradient Diversity and Model Smoothness

Adversarial Transferability is an intriguing property of adversarial exa...
research
06/25/2020

Does Adversarial Transferability Indicate Knowledge Transferability?

Despite the immense success that deep neural networks (DNNs) have achiev...
research
06/06/2023

Quantifying the Variability Collapse of Neural Networks

Recent studies empirically demonstrate the positive relationship between...
research
07/15/2023

Why Does Little Robustness Help? Understanding Adversarial Transferability From Surrogate Training

Adversarial examples (AEs) for DNNs have been shown to be transferable: ...
research
10/16/2022

Non-Transferability in Communication Channels and Tarski's Truth Theorem

This article aims to study transferability issues in communication chann...

Please sign up or login with your details

Forgot password? Click here to reset