Adaptive reward-poisoning attacks against reinforcement learning

X Zhang, Y Ma, A Singla, X Zhu - … Conference on Machine …, 2020 - proceedings.mlr.press
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward $ r_t $ into $ r_t+\delta_t $ at each step, with the goal of forcing the RL …
被引用次数：133 相关文章
[PDF] wisc.edu

[PDF][PDF] Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - pages.cs.wisc.edu
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward rt into rt+ δt at each step, with the goal of forcing the RL agent to learn a …

Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - arXiv e-prints, 2020 - ui.adsabs.harvard.edu
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward $ r_t $ into $ r_t+\delta_t $ at each step, with the goal of forcing the RL …

Adaptive reward-poisoning attacks against reinforcement learning

X Zhang, Y Ma, A Singla, X Zhu - … of the 37th International Conference on …, 2020 - dl.acm.org
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward rt into rt+ δ t at each step, with the goal of forcing the RL agent to learn a …
[PDF] nsf.gov

[PDF][PDF] Adaptive reward-poisoning attacks against reinforcement learning

X Zhang, Y Ma, A Singla, X Zhu - International Conference on Machine …, 2020 - par.nsf.gov
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward rt into rt+ δt at each step, with the goal of forcing the RL agent to learn a …

Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - arXiv preprint arXiv:2003.12613, 2020 - arxiv.org
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward $ r_t $ into $ r_t+\delta_t $ at each step, with the goal of forcing the RL …

Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - … Conference on Machine …, 2020 - proceedings.mlr.press
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward $ r_t $ into $ r_t+\delta_t $ at each step, with the goal of forcing the RL …

[引用][C] Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - 37th International Conference on …, 2020 - pure.mpg.de
Adaptive Reward-Poisoning Attacks against Reinforcement Learning :: MPG.PuRe English
Help Privacy Policy Disclaimer Include files Advanced SearchBrowse START BASKET (0)Tools …
[PDF] nsf.gov

[PDF][PDF] Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - International Conference on Machine …, 2020 - par.nsf.gov
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward rt into rt+ δt at each step, with the goal of forcing the RL agent to learn a …
[PDF] wisc.edu

[PDF][PDF] Adaptive Reward-Poisoning Attacks against Reinforcement Learning

X Zhang, Y Ma, A Singla, X Zhu - pages.cs.wisc.edu
In reward-poisoning attacks against reinforcement learning (RL), an attacker can perturb the
environment reward rt into rt+ δt at each step, with the goal of forcing the RL agent to learn a …