Yes We can: Watermarking Machine Learning Models beyond Classification
| Autor: | Slim Trabelsi, Melek Önen, Sofiane Lounici, Orhan Ermis, Mohamed Njeh |
|---|---|
| Přispěvatelé: | Eurecom [Sophia Antipolis], ANR-19-P3IA-0002,3IA@cote d'azur,3IA Côte d'Azur(2019) |
| Rok vydání: | 2021 |
| Předmět: |
Artificial neural network
Machine translation Contextual image classification Computer science business.industry Process (engineering) [INFO.INFO-IA]Computer Science [cs]/Computer Aided Engineering computer.software_genre Machine learning Data modeling [INFO.INFO-CR]Computer Science [cs]/Cryptography and Security [cs.CR] [STAT.ML]Statistics [stat]/Machine Learning [stat.ML] Robustness (computer science) Reinforcement learning Artificial intelligence business computer Digital watermarking ComputingMilieux_MISCELLANEOUS |
| Zdroj: | CSF CFS 2021, 34th IEEE Computer Security Foundations Symposium CFS 2021, 34th IEEE Computer Security Foundations Symposium, Jun 2021, Dubrovnik, Croatia |
| DOI: | 10.1109/csf51468.2021.00044 |
| Popis: | Since machine learning models have become a valuable asset for companies, watermarking techniques have been developed to protect the intellectual property of these models and prevent model theft. We observe that current watermarking frameworks solely target image classification tasks, neglecting a considerable part of machine learning techniques. In this paper, we propose to address this lack and study the watermarking process of various machine learning techniques such as machine translation, regression, binary image classification and reinforcement learning models. We adapt current definitions to each specific technique and we evaluate the main characteristics of the watermarking process, in particular the robustness of the models against a rational adversary. We show that watermarking models beyond classification is possible while preserving their overall performance. We further investigate various attacks and discuss the importance of the performance metric in the verification process and its impact on the success of the adversary. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|