AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

Provably Efficient Black-Box Action Poisoning Attacks Against Reinforcement Learning

Guanlin Liu, Lifeng Lai

NeurIPS 2021 · Adversarial robustness and security › Black-box and transfer attacks · approach: Query-efficient black-box attacks · 12 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 the α-portion attack, replacing non-target actions with the target action or a worst action under the target policyAny RL agent with sublinear regret in a tabular episodic MDP · Tabular episodic MDP, white-boxexpected cost and loss: E[Cost(K, H)] ≤ E[Loss(K, H)] ≤ R(T)/(α Δ_min)Improved
Estimate target-policy Q-values with importance sampling and explore low-value actions using lower confidence boundsEfficient RL agents with sublinear expected dynamic regret · Tabular episodic MDP, black-boxcost and loss: Cost(K, H) ≤ Loss(K, H) ≤ H(R(T) + 2H^2√(log(1/p)R(T)))/Δ_min + 307SAH^4log(2SAT/p)/Δ_min^2Improved
Use LCB-H instead of the white-box 1/H-portion attackUCB-H, UCB-B, and UCBVI-CH · Periodic 1-d grid worldcumulative attack cost and lossNo clear effect
Use LCB-H instead of the white-box 1/H-portion attackUCB-H, UCB-B, and UCBVI-CH · 4-by-4 grid worldcumulative attack cost and lossNo clear effect

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