The Annals of Statistics

Two-Armed Dirichlet Bandits with Discounting

Manas K. Chattopadhyay

Full-text: Open access


Sequential selections are to be made from two independent stochastic processes, or "arms." At each stage we choose which arm to observe based on past selections and observations. The observations on arm $i$ are conditionally i.i.d. given their marginal distribution $P_i$ which has a Dirichlet process prior with parameter $\alpha_i, i = 1, 2$. Future observations are discounted: at stage $m$, the payoff is $a_m$ times the observation $Z_m$ at that stage. The discount sequence $A_n = (a_1, a_2,\cdots, a_n, 0,0,\cdots)$ is a nonincreasing sequence of nonnegative numbers, where the "horizon" $n$ is finite. The objective is to maximize the total expected payoff $E(\sum^n_1a_iZ_i)$. It is shown that optimal strategies continue with an arm when it yields a sufficiently large observation, one larger than a "break-even observation." This generalizes results of Clayton and Berry, who considered two arms with one arm known and assumed $a_m = 1 \forall m \leq n$.

Article information

Ann. Statist., Volume 22, Number 3 (1994), 1212-1221.

First available in Project Euclid: 11 April 2007

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier


Primary: 62L05: Sequential design
Secondary: 62C10: Bayesian problems; characterization of Bayes procedures

Sequential decisions two-armed bandits one-armed bandits Dirichlet bandits Dirichlet process prior


Chattopadhyay, Manas K. Two-Armed Dirichlet Bandits with Discounting. Ann. Statist. 22 (1994), no. 3, 1212--1221. doi:10.1214/aos/1176325626.

Export citation