AI Paper Atlas

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

Stealthy and Efficient Adversarial Attacks against Deep Reinforcement Learning

Jianwen Sun, Tianwei Zhang, Xiaofei Xie, Lei Ma, Yan Zheng, Kangjie Chen, Yang Liu

AAAI 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Defenses against universal perturbations · 127 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
Use Critical Point Attack to select attack moments using predicted long-term damageA3C · Pongattack steps: 6 steps for Pong and 2 stepsImproved
Use Critical Point Attack to select attack moments using predicted long-term damageA3C · Breakoutattack steps: 5 steps for Breakout and 2 stepsImproved
Use Critical Point Attack to select attack moments using predicted long-term damageDDPG · TORCSattack steps: 80 steps; 1 stepImproved
Use Antagonist Attack to learn attack timing and target actionsA3C · Pongattack steps: 6 steps for Pong and 3 stepsImproved
Use Antagonist Attack to learn attack timing and target actionsA3C · Breakoutattack steps: 5 steps for Breakout and 3 stepsImproved
Use Antagonist Attack to learn attack timing and target actionsPPO · InvertedPendulumattack steps: 2Improved
Use Antagonist Attack to learn attack timing and target actionsPPO · Hopperattack steps: 1Improved
Use Antagonist Attack to learn attack timing and target actionsPPO · HalfCheetahattack steps: 5Improved
Use Antagonist Attack to learn attack timing and target actionsPPO · Walker2dattack steps: 3Improved
Use Critical Point Attack instead of Antagonist AttackA3C · Pong and Breakoutattack steps: Antagonist Attack ... 3 steps; Critical Point Attack (2 steps)Improved

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