Contextual Bandit on Contextual Bandit Theoretical Bounds
1Regret ScalingLinUCB
Evaluation results
| Method | Links | |
|---|---|---|
| LinUCBAssumption on reward=Linear2026.05 | 1 | |
| LinTSAssumption on reward=Linear2026.05 | 3 | |
| LinUCB (feature diversity)Assumption on reward=Linear, Condition=feature diversity assumption2026.05 | 16 | |
| GLM-UCBAssumption on reward=Generalized linear model2026.05 | 17 | |
| GLM-UCB (feature diversity)Assumption on reward=Generalized linear model, Condition=feature diversity assumption2026.05 | 29 | |
| KernelUCBAssumption on reward=Lies in RKHS2026.05 | 38 |