Research map / Bandits and online learning / Online convex optimization and games
Non-stationary Bandits with Knapsacks
Shang Liu, Jiashuo Jiang, Xiaocheng Li
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 naive stationary UCB instead of SW-UCB in the one-dimensional two-armed instance | SW-UCB · Example 1 (d=1) | cumulative reward | No clear effect |
| use naive stationary UCB instead of SW-UCB in the two-dimensional instance | SW-UCB · Example 2 (d=2) | cumulative reward | Hurt |
| use adversarial LagrangeBwK instead of SW-UCB | SW-UCB · Examples 1 and 2 | cumulative reward | Hurt |
| increase W with V fixed | SW-UCB and naive UCB · Example 3 | regret | Hurt |
| increase W with V fixed | LagrangeBwK · Example 3 | regret | No clear effect |
| use SW-UCB instead of LagrangeBwK at small W | SW-UCB · Example 3 | regret | Improved |
| use SW-UCB instead of LagrangeBwK at very large W | SW-UCB · Example 3 | regret | Hurt |
| use SW-UCB instead of LagrangeBwK in slowly varying environments | SW-UCB · Example 4 (fixed W, varying V) | regret | Improved |
| increase V (rapidly changing environment) | SW-UCB · Example 4 | regret | Hurt |
| rapidly changing environment, SW-UCB versus LagrangeBwK | SW-UCB · Example 4 | regret | Improved |
Closest papers by meaning
- Constrained Feedback Learning for Non-Stationary Multi-Armed Bandits (NeurIPS 2025 · similarity 0.92)
- Piecewise-Stationary Bandits with Knapsacks (NeurIPS 2024 · similarity 0.92)
- Non-stationary Bandit Convex Optimization: A Comprehensive Study (NeurIPS 2025 · similarity 0.91)
- No-Regret is not enough! Bandits with General Constraints through Adaptive Regret Minimization (ICML 2025 · similarity 0.91)
- Bandits with Replenishable Knapsacks: the Best of both Worlds (ICLR 2024 · similarity 0.91)
- Tracking Most Significant Shifts in Infinite-Armed Bandits (ICML 2025 · similarity 0.90)
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.
