通过扰动特征实现高效探索，显著降低上下文老虎机的误差上限。
Exploration via Feature Perturbation in Contextual Bandits
- 直接在输入特征上加随机扰动，不依赖参数采样或奖励噪声。
- 理论证明其最坏情况误差为退化的 $\tilde{\mathcal{O}}(d\sqrt{T})$，优于传统方法。
- 计算高效，适合非参数模型和神经网络，兼具理论与实用优势。
我们提出特征扰动，一种针对上下文老虎机的简单而有效的探索策略，该策略直接向特征输入注入随机性，而非对未知参数进行随机化或向奖励添加噪声。令人惊讶的是，该算法在广义线性上下文老虎机中实现了 $\tilde{\mathcal{O}}(d\sqrt{T})$ 的最坏情况后悔界，而无需承担现有随机化老虎机算法常见的 $\tilde{\mathcal{O}}(d^{3/2}\sqrt{T})$ 后悔。由于避免了参数采样，该算法在计算上高效，并自然适用于非参数或神经网络模型。通过实证评估验证了这些优势，表明特征扰动不仅超越现有方法，还统一了强大的实际性能与近最优的后悔保证。
原文摘要 · Abstract (English)
We propose feature perturbation, a simple yet effective exploration strategy for contextual bandits that injects randomness directly into feature inputs, instead of randomizing unknown parameters or adding noise to rewards. Remarkably, this algorithm achieves $\tilde{\mathcal{O}}(d\sqrt{T})$ worst-case regret bound for generalized linear contextual bandits, while avoiding the $\tilde{\mathcal{O}}(d^{3/2}\sqrt{T})$ regret typical of existing randomized bandit algorithms. Because our algorithm eschews parameter sampling, it is both computationally efficient and naturally extends to non-parametric or neural network models. We verify these advantages through empirical evaluations, demonstrating that feature perturbation not only surpasses existing methods but also unifies strong practical performance with the near-optimal regret guarantees.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。