English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Paper

Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning

MPS-Authors
/persons/resource/persons197297

Bhattacharyya,  Apratim
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)

arXiv:1904.01067.pdf
(Preprint), 2MB

Supplementary Material (public)
There is no public supplementary material available
Citation

Salem, A. M. G., Bhattacharyya, A., Backes, M., Fritz, M., & Zhang, Y. (2019). Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning. Retrieved from http://arxiv.org/abs/1904.01067.


Cite as: http://hdl.handle.net/21.11116/0000-0003-EC51-8
Abstract
Machine learning (ML) has progressed rapidly during the past decade and the major factor that drives such development is the unprecedented large-scale data. As data generation is a continuous process, this leads to ML service providers updating their models frequently with newly-collected data in an online learning scenario. In consequence, if an ML model is queried with the same set of data samples at two different points in time, it will provide different results. In this paper, we investigate whether the change in the output of a black-box ML model before and after being updated can leak information of the dataset used to perform the update. This constitutes a new attack surface against black-box ML models and such information leakage severely damages the intellectual property and data privacy of the ML model owner/provider. In contrast to membership inference attacks, we use an encoder-decoder formulation that allows inferring diverse information ranging from detailed characteristics to full reconstruction of the dataset. Our new attacks are facilitated by state-of-the-art deep learning techniques. In particular, we propose a hybrid generative model (BM-GAN) that is based on generative adversarial networks (GANs) but includes a reconstructive loss that allows generating accurate samples. Our experiments show effective prediction of dataset characteristics and even full reconstruction in challenging conditions.