TSEC: A Framework for Online Experimentation under Experimental Constraints
TSEC: A Framework for Online Experimentation under Experimental Constraints
复制标题
TSEC:实验约束下的在线实验框架
DOI:
10.1080/00401706.2022.2125443
复制
发表时间:
2022
期刊:
影响因子:
2.5
通讯作者:
Wu, C. F.
中科院分区:
文献类型:
--
作者:
Mak, Simon;Zhao, Yuanshuo;Hoang, Lavonne;Wu, C. F.
Thompson sampling is a popular algorithm for tackling multi-armed bandit problems, and has been applied in a wide range of applications, from website design to portfolio optimization. In such applications, however, the number of choices (or arms)Ncan be large, and the data needed to make adaptive decisions require expensive experimentation. One is then faced with the constraint of experimenting on only a small subset ofarms within each time period, which poses a problem for traditional Thompson sampling. We propose a new Thompson Sampling under Experimental Constraints (TSEC) method, which addresses this so-called “arm budget constraint.” TSEC makes use of a Bayesian interaction model with effect hierarchy priors, to model correlations between rewards on different arms. This fitted model is then integrated within Thompson sampling, to jointly identify a good subset of arms for experimentation and to allocate resources over these arms. We demonstrate the effectiveness of TSEC in two applications with arm budget constraints. The first is a simulated website optimization study, where TSEC shows considerable improvements over industry benchmarks. The second is a portfolio optimization application on industry-based exchange-traded funds, where TSEC provides more consistent and greater wealth accumulation over standard investment strategies.
登录
查看更多内容
DOI:
--
发表时间:
2014-11
期刊:
ArXiv
影响因子:
--
作者:
Tor Lattimore;R. Munos
通讯作者:
Tor Lattimore;R. Munos
DOI:
--
发表时间:
2006-12
期刊:
J. Mach. Learn. Res.
影响因子:
--
作者:
Eyal Even-Dar;Shie Mannor;Y. Mansour
通讯作者:
Eyal Even-Dar;Shie Mannor;Y. Mansour
DOI:
10.1145/1273496.1273587
发表时间:
2007-06
期刊:
--
影响因子:
--
作者:
Sandeep Pandey;Deepayan Chakrabarti;D. Agarwal
通讯作者:
Sandeep Pandey;Deepayan Chakrabarti;D. Agarwal
影响因子:
2.5
作者:
Jason L. Loeppky
通讯作者:
Jason L. Loeppky
影响因子:
4.8
作者:
Domingo, C;Gavaldà, R;Watanabe, O
通讯作者:
Watanabe, O