AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Spatiotemporally Constrained Action Space Attacks on Deep Reinforcement Learning Agents

Xian Yeow Lee, Sambit Ghadai, Kai Liang Tan, Chinmay Hegde, Soumik Sarkar

AAAI 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Attacks and defenses in sequential settings · 62 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add look-ahead planning that allocates attack budget across time, rather than using myopic per-step attacksPPO · Lunar Landercumulative rewardImproved
Add look-ahead planning that allocates attack budget across time, rather than using myopic per-step attacksPPO · Bipedal Walkercumulative rewardImproved
Add look-ahead planning that allocates attack budget across time, rather than using myopic per-step attacksDoubleDQN · Lunar Lander and Bipedal Walkercumulative rewardImproved
Add look-ahead planning that allocates attack budget across time, rather than using myopic per-step attacksPPO · Hopper, Half-Cheetah, and Walkercumulative rewardImproved
Increase LAS budget (ablation)PPO · Lunar Landercumulative rewardHurt
Increase LAS budget beyond the point where attacks saturate (ablation)LAS · Lunar Landercumulative reward reduction versus MASNo clear effect
Use a longer horizon with the same overall budget (ablation)LAS · Lunar Landercumulative rewardMixed
Use ℓ2 rather than ℓ1 temporal projection to distribute attacks more frequently (ablation)LAS · Lunar Landercumulative rewardImproved
Use ℓ1 rather than ℓ2 temporal projection to concentrate attacks at fewer time points (ablation)LAS · Lunar Landercumulative rewardHurt
Increase budget with ℓ1 spatial projection at horizon 5 beyond B=3 (ablation)LAS · Lunar Landercumulative rewardNo clear effect
Use action-dimension-aware LAS perturbations, concentrating more attack on the vulnerable Up-Down action (ablation)PPO · Lunar Landercumulative rewardImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy