Satisficing in Time-Sensitive Bandit Learning
This addresses the issue of time sensitivity in bandit learning for scenarios where optimal actions require excessive information, offering a practical solution for applications with computational or data constraints.
The paper tackles the problem of time-sensitive bandit learning by proposing satisficing Thompson sampling to learn near-optimal actions with less information, demonstrating arbitrarily large benefits over Thompson sampling in linear and infinite-armed bandits.
Much of the recent literature on bandit learning focuses on algorithms that aim to converge on an optimal action. One shortcoming is that this orientation does not account for time sensitivity, which can play a crucial role when learning an optimal action requires much more information than near-optimal ones. Indeed, popular approaches such as upper-confidence-bound methods and Thompson sampling can fare poorly in such situations. We consider instead learning a satisficing action, which is near-optimal while requiring less information, and propose satisficing Thompson sampling, an algorithm that serves this purpose. We establish a general bound on expected discounted regret and study the application of satisficing Thompson sampling to linear and infinite-armed bandits, demonstrating arbitrarily large benefits over Thompson sampling. We also discuss the relation between the notion of satisficing and the theory of rate distortion, which offers guidance on the selection of satisficing actions.