Related papers: Tactics of Adversarial Attack on Deep Reinforcemen…
Current neural-network-based classifiers are susceptible to adversarial examples. The most empirically successful approach to defending against such adversarial examples is adversarial training, which incorporates a strong self-attack…
Recent research showed that deep neural networks are highly sensitive to so-called adversarial perturbations, which are tiny perturbations of the input data purposely designed to fool a machine learning classifier. Most classification…
Recent developments in deep reinforcement learning have enabled the creation of agents for solving a large variety of games given a visual input. These methods have been proven successful for 2D games, like the Atari games, or for simple…
We study reward poisoning attacks on online deep reinforcement learning (DRL), where the attacker is oblivious to the learning algorithm used by the agent and the dynamics of the environment. We demonstrate the intrinsic vulnerability of…
In this paper we explore how actor-critic methods in deep reinforcement learning, in particular Asynchronous Advantage Actor-Critic (A3C), can be extended with agent modeling. Inspired by recent works on representation learning and…
Reinforcement learning agents in complex game environments often suffer from sparse rewards, training instability, and poor sample efficiency. This paper presents a hybrid training approach that combines offline imitation learning with…
In this paper, we propose a defence strategy to improve adversarial robustness by incorporating hidden layer representation. The key of this defence strategy aims to compress or filter input information including adversarial perturbation.…
Reinforcement learning agents learn by encouraging behaviours which maximize their total reward, usually provided by the environment. In many environments, however, the reward is provided after a series of actions rather than each single…
In recent years, many efforts have demonstrated that modern machine learning algorithms are vulnerable to adversarial attacks, where small, but carefully crafted, perturbations on the input can make them fail. While these attack methods are…
In targeted poisoning attacks, an attacker manipulates an agent-environment interaction to force the agent into adopting a policy of interest, called target policy. Prior work has primarily focused on attacks that modify standard MDP…
Adversarial training aims to defend against adversaries: malicious opponents whose sole aim is to harm predictive performance in any way possible. This presents a rather harsh perspective, which we assert results in unnecessarily…
Recent studies have shown that deep reinforcement learning agents are vulnerable to small adversarial perturbations on the agent's inputs, which raises concerns about deploying such agents in the real world. To address this issue, we…
Single-agent reinforcement learning algorithms in a multi-agent environment are inadequate for fostering cooperation. If intelligent agents are to interact and work together to solve complex problems, methods that counter non-cooperative…
Deep learning classifiers are known to be inherently vulnerable to manipulation by intentionally perturbed inputs, named adversarial examples. In this work, we establish that reinforcement learning techniques based on Deep Q-Networks (DQNs)…
We propose a new perspective on adversarial attacks against deep reinforcement learning agents. Our main contribution is CopyCAT, a targeted attack able to consistently lure an agent into following an outsider's policy. It is pre-computed,…
Training intelligent agents through reinforcement learning is a notoriously unstable procedure. Massive parallelization on GPUs and distributed systems has been exploited to generate a large amount of training experiences and consequently…
Autonomous cars are well known for being vulnerable to adversarial attacks that can compromise the safety of the car and pose danger to other road users. To effectively defend against adversaries, it is required to not only test autonomous…
Reinforcement learning policies based on deep neural networks are vulnerable to imperceptible adversarial perturbations to their inputs, in much the same way as neural network image classifiers. Recent work has proposed several methods to…
Adversarial training (AT) is among the most effective techniques to improve model robustness by augmenting training data with adversarial examples. However, most existing AT methods adopt a specific attack to craft adversarial examples,…
We pose an active perception problem where an autonomous agent actively interacts with a second agent with potentially adversarial behaviors. Given the uncertainty in the intent of the other agent, the objective is to collect further…