CBT: Confidence Bound Target Algorithm

The Confidence Bound Target (CBT) algorithm is designed for infinite arms bandit problem. It is shown that CBT algorithm achieves the regret lower bound for general reward distributions. Reference: Hock Peng Chan and Shouri Hu (2018) <arXiv:1805.11793>.

Getting started

Package details

AuthorHock Peng Chan and Shouri Hu
MaintainerShouri Hu <e0054325@u.nus.edu>
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the CBT package in your browser

Any scripts or data that you put into this service are public.

CBT documentation built on May 2, 2019, 8:17 a.m.