AI Paper Atlas

Research map / Bandits and online learning / Contextual bandits and Thompson sampling

Adversarial Attacks on Linear Contextual Bandits

Evrard Garcelon, Baptiste Roziere, Laurent Meunier, Jean Tarbouriech, Olivier Teytaud, Alessandro Lazaric, Matteo Pirotta

NeurIPS 2020 · Bandits and online learning › Contextual bandits and Thompson sampling · approach: Practical contextual bandit extensions · 18 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
Apply ContextualConic context attack to LinUCBLinUCB · Synthetic, Jester, and MovieLenstarget-set selection rate: 28.91% -> 98.55%; 26.59% -> 98.36%; 31.13% -> 99.61%Improved
Apply ContextualConic context attack to ε-greedyε-greedy · Synthetic, Jester, and MovieLenstarget-set selection rate: 25.7% -> 89.71%; 25.85% -> 99.85%; 31.78% -> 99.92%Improved
Apply ContextualConic context attack to LinTSLinTS · Synthetictarget-set selection rate: 27.2% -> 30.93%No clear effect
Apply ContextualConic context attack to LinTSLinTS · Jester and MovieLenstarget-set selection rate: 26.10% -> 97.26%; 33.24% -> 98.82%Improved
Use full rather than relaxed optimization for the offline LinUCB context attackLinUCB offline context attack · Synthetic and MovieLenstarget-set selection rate: 13.76% -> 88.30%; 4.09% -> 99.99%Improved
Use full rather than relaxed optimization for the offline LinUCB context attackLinUCB offline context attack · Jestertarget-set selection rate: 97.81% -> 99.98%Improved
Apply full offline context attackε-greedy · Synthetic, Jester, and MovieLenstarget-set selection rate: 0.01% -> 99.98%; 0.00% -> 99.95%; 0.03% -> 99.97%Improved
Apply relaxed offline context attackLinTS · Synthetic, Jester, and MovieLenstarget-set selection rate: 0.02% -> 18.21%; 0.01% -> 80.48%; 0.05% -> 5.56%Mixed
Apply CACE reward attackLinUCB · Synthetic, Jester, and Movilenstarget arm draws: 86, 731.6 -> 996, 238.6; 23, 548.16 -> 921, 083.69; 25, 017.31 -> 944, 721.28Improved
Use stationary CACE rather than CACE reward attackLinUCB · Synthetic, Jester, and Movilenstarget arm draws: 996, 238.6 -> 995, 578.88; 921, 083.69 -> 862, 095.67; 944, 721.28 -> 931, 531.6Mixed
Attack rewards with CACEcontextual bandit algorithms · Synthetic datatarget-arm selection rate: more than 99.4%; more than 97.8%Improved
Attack rewards with CACEcontextual bandit algorithms · Jester and MovieLenstarget-arm selection rate: more than 94.0%; more than 77.0%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