Bounded Myopic Adversaries for Deep Reinforcement Learning Agents

2021-01-01 · Ezgi Korkmaz, Henrik Sandberg, Gyorgy Dan

Adversarial attacks against deep neural networks have been widely studied. Adversarial examples for deep reinforcement learning (DeepRL) have significant security implications, due to the deployment of these algorithms in many application domains. In this work we formalize an optimal myopic adversary for deep reinforcement learning agents. Our adversary attempts to find a bounded perturbation of the state which minimizes the value of the action taken by the agent. We show with experiments in various games in the Atari environment that our attack formulation achieves significantly larger impact as compared to the current state-of-the-art. Furthermore, this enables us to lower the bounds by several orders of magnitude on the perturbation needed to efficiently achieve significant impacts on DeepRL agents.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Reinforcement Learningreinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Guarantees for Epsilon-Greedy Reinforcement Learning with Function Approximation

2022-06-19 · Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari 외

Myopic exploration policies such as epsilon-greedy, softmax, or Gaussian noise fail to explore efficiently in some reinforcement learning tasks and yet, they perform well in many others. In fact, in practice, they are of…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Multi-Task Federated Reinforcement Learning with Adversaries

2021-03-11 · Aqeel Anwar, Arijit Raychowdhury

Reinforcement learning algorithms, just like any other Machine learning algorithm pose a serious threat from adversaries. The adversaries can manipulate the learning algorithm resulting in non-optimal policies. In this p…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Resilient Dynamic Average Consensus based on Trusted agents

2023-03-14 · Shamik Bhattacharyya, Rachel Kalpana Kalaimani

In this paper, we address the discrete-time dynamic average consensus (DAC) of a multi-agent system in the presence of adversarial attacks. The adversarial attack is considered to be of Byzantine type, which compromises …

Adversarial Attack

Multi-Agent Consensus Subject to Communication and Privacy Constraints

2021-02-21 · Dipankar Maity, Panagiotis Tsiotras

We consider a multi-agent consensus problem in the presence of adversarial agents. The adversaries are able to listen to the inter-agent communications and try to estimate the state of the agents. The agents have a limit…

Quantization

Shared Control with Black Box Agents using Oracle Queries

2024-10-25 · Inbal Avraham, Reuth Mirsky

Shared control problems involve a robot learning to collaborate with a human. When learning a shared control policy, short communication between the agents can often significantly reduce running times and improve the sys…