This paper explores a new form of the linear bandit problem in which the algorithm receives the usual stochastic rewards as well as stochastic feedback about which features are relevant to the rewards, the latter feedback being the novel aspect. The focus of this paper is the development of new theory and algorithms for linear bandits with feature feedback which can achieve regret over time horizon T that scales like k√T, without prior knowledge of which features are relevant nor the number k of relevant features. In comparison, the regret of traditional linear bandits is d√T, where d is the total number of (relevant and irrelevant) features, so the improvement can be dramatic if k ≪ d. The computational complexity of the algorithm is proportional to k rather than d, making it much more suitable for real-world applications compared to traditional linear bandits. We demonstrate the performance of the algorithm with synthetic and real human-labeled data.
Linear Bandits with Feature Feedback
