DNN Fingerprint Removal Attack

DNN fingerprint removal attacks target methods used to verify ownership of deep learning models, aiming to eliminate embedded "fingerprints" that identify the model's origin. Current research focuses on developing both robust fingerprinting techniques (e.g., leveraging neuron functionality analysis, sample correlations, or universal adversarial perturbations) and effective removal attacks (e.g., min-max bilevel optimization). This area is crucial for protecting intellectual property in the rapidly expanding field of deep learning and has significant implications for the security and trustworthiness of AI systems.

Papers

July 4, 2024

Protecting Deep Learning Model Copyrights with Adversarial Example-Free Reuse Detection
Xiaokun Luan, Xiyue Zhang, Jingyi Wang, Meng Sun
Model Reuse Code Clone Detection DNN Fingerprint Removal Attack

August 23, 2023

RemovalNet: DNN Fingerprint Removal Attacks
Hongwei Yao, Zheng Li, Kunzhe Huang, Jian Lou, Zhan Qin, Kui Ren
Active Removal DNN Fingerprint Removal Attack

July 27, 2023

FLARE: Fingerprinting Deep Reinforcement Learning Agents using Universal Adversarial Masks
Buse G. A. Tekgul, N. Asokan
Deep Reinforcement Learning Adversarial Example Retrieval Augmented Generation Adversarial Mask DNN Fingerprint Removal Attack Adversary Knowledge

June 2, 2023

Quantifying Sample Anonymity in Score-Based Generative Models with Adversarial Fingerprinting
Mischa Dombrowski, Bernhard Kainz
Generative Model Score Based Generative Training Image Training Sample DNN Fingerprint Removal Attack Private Image

April 17, 2023

GrOVe: Ownership Verification of Graph Neural Networks using Embeddings
Asim Waheed, Vasisht Duddu, N. Asokan
Graph Neural Network Jina Embeddings Graph Structured Data Model Extraction Attack Graph Node Plant Leaf Ownership Verification DNN Fingerprint Removal Attack

November 24, 2022

DeepTaster: Adversarial Perturbation-Based Fingerprinting to Identify Proprietary Dataset Use in Deep Neural Networks
Seonhye Park, Alsharif Abuadbba, Shuo Wang, Kristen Moore, Yansong Gao, Hyoungshick Kim, Surya Nepal
Deep Neural Network Adversarial Image Model Stealing Attack DNN Fingerprint Removal Attack

October 21, 2022

Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural Networks
Jiyang Guan, Jian Liang, Ran He
Adversarial Example Full Model Correlation Analysis Suspect Model DNN Fingerprint Removal Attack

February 17, 2022

Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations
Zirui Peng, Shaofeng Li, Guoxing Chen, Cheng Zhang, Haojin Zhu, Minhui Xue
DNN Model Model Extraction Attack Intellectual Property Protection Suspect Model DNN Fingerprint Removal Attack

DNN Fingerprint Removal Attack

Papers

Protecting Deep Learning Model Copyrights with Adversarial Example-Free Reuse Detection

RemovalNet: DNN Fingerprint Removal Attacks

FLARE: Fingerprinting Deep Reinforcement Learning Agents using Universal Adversarial Masks

Quantifying Sample Anonymity in Score-Based Generative Models with Adversarial Fingerprinting

GrOVe: Ownership Verification of Graph Neural Networks using Embeddings

DeepTaster: Adversarial Perturbation-Based Fingerprinting to Identify Proprietary Dataset Use in Deep Neural Networks

Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural Networks

Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations