Determining optimal actions requires an appropriate balance of exploration and exploitation at each stage. In the traditional setting, actions which maximize the cumulative expected reward are deemed to be optimal. The MAB framework has seen many practical applications in a wide variety of fields like healthcare, finance, machine learning and telecommunication to name a few [Bouneffouf and Rish, 2019]. Recent literature has extended the bandits framework with alternative objectives such as Risk-Averse Multiarmed Bandits (RA-MAB) and Budgeted Multiarmed Bandits (B-MAB), which broaden the scope of applications of bandits models. The RA-MAB are concerned with the risk of rewards [Sani et al., 2012] and the B-MAB with a cost associated with each action that depletes a finite budget [Xia et al., 2017].

We derive an instantaneous (per-round) data-dependent regret bound for stochastic multiarmed bandits with side information (also known as contextual bandits).

We derive an instantaneous (per-round) data-dependent regret bound for stochastic multiarmed bandits with side information (also known as contextual bandits). The scaling of our regret bound with the number of states (contexts) N goes as \sqrt{N I_{\rho_t}(S;A)}, where I_{\rho_t}(S;A) is the mutual information between states and actions (the side information) used by the algorithm at round t . If the algorithm uses all the side information, the regret bound scales as \sqrt{N \ln K}, where K is the number of actions (arms). However, if the side information I_{\rho_t}(S;A) is not fully used, the regret bound is significantly tighter. In the extreme case, when I_{\rho_t}(S;A) 0, the dependence on the number of states reduces from linear to logarithmic.

We derive an instantaneous (per-round) data-dependent regret bound for stochastic multiarmed bandits with side information (also known as contextual bandits). The scaling of our regret bound with the number of states (contexts) $N$ goes as $\sqrt{N I_{\rho_t}(S;A)}$, where $I_{\rho_t}(S;A)$ is the mutual information between states and actions (the side information) used by the algorithm at round $t$. If the algorithm uses all the side information, the regret bound scales as $\sqrt{N \ln K}$, where $K$ is the number of actions (arms). However, if the side information $I_{\rho_t}(S;A)$ is not fully used, the regret bound is significantly tighter. In the extreme case, when $I_{\rho_t}(S;A) 0$, the dependence on the number of states reduces from linear to logarithmic. Our analysis allows to provide the algorithm large amount of side information, let the algorithm to decide which side information is relevant for the task, and penalize the algorithm only for the side information that it is using de facto.

We study a setting in which a learner faces a sequence of A/B tests and has to make as many good decisions as possible within a given amount of time. Each A/B test $n$ is associated with an unknown (and potentially negative) reward $\mu_n \in [-1,1]$, drawn i.i.d. from an unknown and fixed distribution. For each A/B test $n$, the learner sequentially draws i.i.d. samples of a $\{-1,1\}$-valued random variable with mean $\mu_n$ until a halting criterion is met. The learner then decides to either accept the reward $\mu_n$ or to reject it and get zero instead. We measure the learner's performance as the sum of the expected rewards of the accepted $\mu_n$ divided by the total expected number of used time steps (which is different from the expected ratio between the total reward and the total number of used time steps). We design an algorithm and prove a data-dependent regret bound against any set of policies based on an arbitrary halting criterion and decision rule. Though our algorithm borrows ideas from multiarmed bandits, the two settings are significantly different and not directly comparable. In fact, the value of $\mu_n$ is never observed directly in our setting---unlike rewards in stochastic bandits. Moreover, the particular structure of our problem allows our regret bounds to be independent of the number of policies.

We derive an instantaneous (per-round) data-dependent regret bound for stochastic multiarmed bandits with side information (also known as contextual bandits). The scaling of our regret bound with the number of states (contexts) $N$ goes as $\sqrt{N I_{\rho_t}(S;A)}$, where $I_{\rho_t}(S;A)$ is the mutual information between states and actions (the side information) used by the algorithm at round $t$. If the algorithm uses all the side information, the regret bound scales as $\sqrt{N \ln K}$, where $K$ is the number of actions (arms). However, if the side information $I_{\rho_t}(S;A)$ is not fully used, the regret bound is significantly tighter. In the extreme case, when $I_{\rho_t}(S;A) = 0$, the dependence on the number of states reduces from linear to logarithmic. Our analysis allows to provide the algorithm large amount of side information, let the algorithm to decide which side information is relevant for the task, and penalize the algorithm only for the side information that it is using de facto. We also present an algorithm for multiarmed bandits with side information with computational complexity that is a linear in the number of actions.