BODAME: Bilevel optimization for defense against model extraction
Y Mori, A Nitanda, A Takeda - arXiv preprint arXiv:2103.06797, 2021 - arxiv.org
Model extraction attacks have become serious issues for service providers using machine
learning. We consider an adversarial setting to prevent model extraction under the …
learning. We consider an adversarial setting to prevent model extraction under the …
[PDF][PDF] BODAME: Bilevel Optimization for Defense Against Model Extraction
Y Mori, A Nitanda, A Takeda - arXiv preprint arXiv:2103.06797, 2021 - researchgate.net
Abstract Model extraction attacks have become serious issues for service providers using
machine learning. We consider an adversarial setting to prevent model extraction under the …
machine learning. We consider an adversarial setting to prevent model extraction under the …
BODAME: Bilevel Optimization for Defense Against Model Extraction
Y Mori, A Nitanda, A Takeda - arXiv e-prints, 2021 - ui.adsabs.harvard.edu
Abstract Model extraction attacks have become serious issues for service providers using
machine learning. We consider an adversarial setting to prevent model extraction under the …
machine learning. We consider an adversarial setting to prevent model extraction under the …