Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Fatigue-Aware Bandits for Dependent Click Models
Junyu Cao, Wei Sun, Zuo-Jun (Max) Shen, Markus Ettl
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use FA-DCM instead of explore-then-exploit | Fatigue-aware DCM bandit · Taobao data, T=10,000 | average regret: 1553.41 vs 1677.37 | Improved |
| Use FA-DCM instead of explore-then-exploit | Fatigue-aware DCM bandit · Simulated data, T=10,000 | average regret: 29.02% higher | Improved |
| Decrease g from 0.95 to 0.85 | FA-DCM-P · Known discount factor, T=10,000 | average regret: 307.52 -> 277.77 | Improved |
| Decrease g from 0.85 to 0.75 | FA-DCM-P · Known discount factor, T=10,000 | average regret: 277.77 -> 265.73 | Improved |
| Increase the discounting effect from f(r)=exp(-0.1(r-1)) to f(r)=exp(-0.15(r-1)) | FA-DCM · Unknown discount factor, g=0.85, T=10,000 | average regret: 1237.82 -> 1178.24 | Improved |
| Decrease g from 0.85 to 0.75 | FA-DCM · Unknown discount factor, f(r)=exp(-0.1(r-1)), T=10,000 | average regret: 1237.82 -> 991.97 | Improved |
| Learn the discount factor f in addition to item relevance | FA-DCM compared with FA-DCM-P · Simulated experiments | average regret | Hurt |
Closest papers by meaning
- Impatient Bandits: Optimizing for the Long-Term Without Delay (JMLR 2026 · similarity 0.88)
- DCM Bandits: Learning to Rank with Multiple Clicks (ICML 2016 · similarity 0.88)
- Dynamic Learning of Sequential Choice Bandit Problem under Marketing Fatigue (AAAI 2019 · similarity 0.88)
- Fighting Boredom in Recommender Systems with Linear Reinforcement Learning (NeurIPS 2018 · similarity 0.88)
- Bandits Meet Mechanism Design to Combat Clickbait in Online Recommendation (ICLR 2024 · similarity 0.87)
- Ready for You When You Are Back: Content-Driven Session-Based Recommendation for Continuity of Experience (AAAI 2025 · similarity 0.85)
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.
