Copyright Protection Studies in Deep Learning
Here's an ongoing list to summarize literature about copyright protection in deep learning, including the copyright of models and data.
And we would greatly appreciate your contributions to expand this list! ✧٩(^ω^)و✧
Year | Title | Copyright Subject | Task Type | Method | Authors | Publisher | 🔗 |
---|---|---|---|---|---|---|---|
2019 | BlackMarks: Blackbox Multibit Watermarking for Deep Neural Networks | Model | Image Encoding | Watermarking | Huili Chen, Bita Darvish Rouhani, and Farinaz Koushanfar | ArXiv | |
2020 | Membership Encoding for Deep Learning | Model | Classification | Membership Inference | Congzheng Song, and Reza Shokri | AsiaCCS | |
2020 | Towards Probabilistic Verification of Machine Unlearning | Data | Classification | Dirty-label Backdoor Attack | David M. Sommer, Liwei Song, Sameer Wagh, and Prateek Mittal | ArXiv | |
2021 | Deep Neural Network Fingerprinting by Conferrable Adversarial Examples | Model | Classification | Adversarial Training; Transfer Learning | Nils Lukas, Yuxuan Zhang, and Florian Kerschbaum | ICLR | pdf & code |
2022 | Defending against Model Stealing via Verifying Embedded External Features | Model | Classification | Dirty-label Backdoor Attack; Hypothesis Testing | Li Yiming, Zhu Linghui, Jia Xiaojun, Jiang Yong, Xia Shu-Tao, and Cao Xiaochun | AAAI | pdf & code |
2022 | Your Model Trains on My Data? Protecting Intellectual Property of Training Data via Membership Fingerprint Authentication | Model & Data | Classification | Membership Inference | Gaoyang Liu, Tianlong Xu, Xiaoqiang Ma, and Chen Wang | TIFS | |
2022 | Deep Model Intellectual Property Protection via Deep Watermarking | Model | Classification | Watermarking; Steganography | Jie Zhang, Dongdong Chen, Jing Liao, Weiming Zhang, Huamin Feng, Gang Hua, and Nenghai Yu | TPAMI | pdf & code |
2022 | Data Isotopes for Data Provenance in DNNs | Data | Classification | Watermarking | Emily Wenger, Xiuyu Li, Ben Y. Zhao, and Vitaly Shmatikov | Arxiv | |
2022 | Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models | Model | Classification | Testing | Jialuo Chen, Jingyi Wang, Tinglan Peng, Youcheng Sun, Peng Cheng, Shouling Ji, Xingjun Ma, Bo Li, and Dawn Song | S&P | pdf & code |
2022 | Non-Transferable Learning: A New Approach for Model Ownership Verification and Applicability Authorization | Model | Classification | Adversarial Training; Domain Shift; | Lixu Wang, Shichao Xu, Ruiqi Xu, Xiao Wang, and Qi Zhu | ICLR | pdf & code |
2022 | Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protection | Dataset | Classification | Dirty-&Clean-label Backdoor Attack; Adversarial Training; Hypothesis Testing | Yiming Li, Yang Bai, Yong Jiang, Yong Yang, Shu-Tao Xia, and Bo Li | NeurIPS | pdf & code |
2023 | Did You Train on My Dataset? Towards Public Dataset Protection with Clean-Label Backdoor Watermarking | Dataset | Classification | Clean-label Backdoor Attack; Hypothesis Testing | Ruixiang Tang, Qizhang Feng, Ninghao Liu, Fan Yang, and Xia Hu | arXiv | pdf & code |
2023 | Black-Box Dataset Ownership Verification via Backdoor Watermarking | Dataset | Classification | Dirty-label Backdoor Attack; Hypothesis Testing | Yiming Li, Mingyan Zhu, Xue Yang, Yong Jiang, Tao Wei, and Shu-Tao Xia | TIFS | pdf & code |
2023 | Domain Watermark: Effective and Harmless Dataset Copyright Protection is Closed at Hand | Dataset | Classification | Adversarial Training; Domain Shift; Hypothesis Testing | Junfeng Guo, Yiming Li, Lixu Wang, Shu-Tao Xia, Heng Huang, Cong Liu, and Bo Li | NeurIPS | pdf & code |
What's the difference between Ownership and Copyright in deep learning?