This paper gives concentration results around the mean of the regret for variants of UCB algorithm in multi-armed stochastic bandits. However, I could not find any similar results for Thompson sampling, where most of the results are only in expectation. Is there any paper which studies concentration of the regret in the case of Thompson sampling ?

  • $\begingroup$ How about writing to the authors directly and reporting back here? I'd be curious to know... $\endgroup$ – Aryeh May 20 '15 at 16:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.