Fetching the paper…

Reading the bibliography…

2019

Adversarial Exploitation of Policy Imitation

Behzadan, Vahid, Hsu, William

Original
PDF
Post on X

Understand

This paper investigates a class of attacks targeting the confidentiality aspect of security in Deep Reinforcement Learning (DRL) policies.

Recent research have established the vulnerability of supervised machine learning models (e.g., classifiers) to model extraction attacks.
Such attacks leverage the loosely-restricted ability of the attacker to iteratively query the model for labels, thereby allowing for the forging of a labeled dataset which can be used to train a replica of the original model.
In this work, we demonstrate the feasibility of exploiting imitation learning techniques in launching model extraction attacks on DRL agents.

Adversarial Exploitation of Policy Imitation · Around

Built on

Boosted bellman residual minimization handling expert demonstrations
Bilal Piot, Matthieu Geist, and Olivier Pietquin · 2014
Earlier work this paper cites.
Delving into transferable adversarial examples and black-box attacks
Original
Yanpei Liu, Xinyun Chen, Chang Liu, and Dawn Song · 2016
Earlier work this paper cites.

Similar

Stealing machine learning models via prediction apis
Florian Tramèr, Fan Zhang, Ari Juels, Michael K Reiter, and Thomas Ristenpart · 2016
Cited alongside, same era.
Vulnerability of deep reinforcement learning to policy induction attacks
Vahid Behzadan and Arslan Munir · 2017
Cited alongside, same era.

Then

Imitation learning: A survey of learning methods
Ahmed Hussein, Mohamed Medhat Gaber, Eyad Elyan, and Chrisina Jayne · 2017
Later among the works it cites.
Deep q-learning from demonstrations
Todd Hester, Matej Vecerik, Olivier Pietquin, Marc Lanctot, Tom Schaul, Bilal Piot, Dan Horgan, John Quan, Andrew Sendonaris, Ian Osband, et al · 2018
Later among the works it cites.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…